0130 points
Safety
Can an agent verify identity, permissions, and intent before it takes a destructive or irreversible action?
Unconfirmed writes · permission preflights
MCP agent-readiness scorecard
Your tools work.
Can an agent tell?
Gradable scans your MCP server and tells you, with a real number, whether an AI agent can actually use it correctly — or is silently guessing wrong.
- Real generated tasks
- Independent judge CLIs
- Actionable fixes
“The function returned 200” is not the same as “the agent understood the function.”
Gradable measures the interface your model actually sees.
The score
Four signals.
One honest number.
Gradable examines the tool surface, runs generated tasks through real agent workflows, and turns the evidence into a weighted score out of 100. It measures agent usability—not popularity, uptime, or stars.
Every formula, weight, and threshold behind these numbers is published. Read how we score, or take issue with it in the open spec on GitHub —including the report format any scanner can emit.
0230 points
Legibility
Do names, descriptions, parameters, and return values tell the model exactly what each tool does?
Description defects · tool confusability
0320 points
Economics
How much context does the server consume before an agent has done any useful work—and is that cost justified?
Definition tokens · context overhead
0420 points
Discoverability
Given a real task, can an agent find and select the correct tool without guessing among near-duplicates?
Selection accuracy · task coverage
How it works
Static schemas meet dynamic judgment.
A useful score cannot come from counting fields alone. Gradable combines structural analysis with generated tasks and CLI-based judges, then identifies the highest-impact change you can make.
- 01Ingest the server
Capture the tools and the interface an agent receives.
- 02Generate real tasks
Test selection and behavior against representative intent.
- 03Judge the outcomes
Score safety, clarity, cost, and discoverability.
- 04Prioritize the fix
Turn the evidence into a scorecard your team can use.
Public leaderboard
The score is the start
of the conversation.
473 completed assessments
Latest scan Aug 4, 2026
Search includes package names and target IDs.
| Rank | Server | Score | Grade | Safety | Legibility | Economics | Discovery | Scanned |
|---|---|---|---|---|---|---|---|---|
| 01 | apify/mcp-server-rag-web-browser (@apify/mcp-server-rag-web-browser) apify-mcp-server-rag-web-browser · v0.1.0 | 100.0/100 | A | 30.0/30 | 30.0/30 | 20.0/20 | 20.0/20 | |
Why this scorePreserve the current agent-ready contractSelection accuracy 100%, destructive-action safety rate 100% (baseline only -- no rewrite pass applied).
Highest-impact fix
View full report
Keep the pinned task set and judge tuple in CI so tool-surface or model changes cannot silently regress this score. |
||||||||
| 02 | zhsama/duckduckgo-mcp-server (duckduckgo-mcp-server) duckduckgo-mcp-server · v0.1.2 | 100.0/100 | A | 30.0/30 | 30.0/30 | 20.0/20 | 20.0/20 | |
Why this scorePreserve the current agent-ready contractSelection accuracy 100%, destructive-action safety rate 100% (baseline only -- no rewrite pass applied).
Highest-impact fix
View full report
Keep the pinned task set and judge tuple in CI so tool-surface or model changes cannot silently regress this score. |
||||||||
| 03 | contextstream/mcp-server (@contextstream/mcp-server) contextstream-mcp-server · v0.4.81 | 99.8/100 | A | 30.0/30 | 29.8/30 | 20.0/20 | 20.0/20 | |
Why this scoreRewrite 1 underspecified tool descriptionSelection accuracy 100→100, token cost up 2%, unconfirmed writes 100%→100%.
Highest-impact fix
View full report
State each tool's behavior, explain every parameter, and declare return semantics, prioritizing tools exercised by failed selection tasks. |
||||||||
| 04 | kimtaeyoon83/mcp-server-youtube-transcript (@kimtaeyoon83/mcp-server-youtube-transcript) kimtaeyoon83-mcp-server-youtube-transcript · v0.1.0 | 99.8/100 | A | 30.0/30 | 29.8/30 | 20.0/20 | 20.0/20 | |
Why this scoreRewrite 1 underspecified tool descriptionSelection accuracy 100→100, token cost up 3%, unconfirmed writes 100%→100%.
Highest-impact fix
View full report
State each tool's behavior, explain every parameter, and declare return semantics, prioritizing tools exercised by failed selection tasks. |
||||||||
| 05 | serpdive/serpdive-mcp (serpdive-mcp) serpdive-mcp · v0.2.1 | 99.7/100 | A | 30.0/30 | 30.0/30 | 20.0/20 | 19.7/20 | |
Why this scoreMake target tools discoverable on the first callSelection accuracy 100%, destructive-action safety rate 100% (baseline only -- no rewrite pass applied).
Highest-impact fix
View full report
Clarify tool names, decision boundaries, and required argument schemas so an agent can choose and construct the target call without exploratory steps. |
||||||||
| 06 | keenableai/keenable-mcp (@keenable/mcp) keenable-mcp · v0.1.1 | 99.7/100 | A | 30.0/30 | 30.0/30 | 20.0/20 | 19.7/20 | |
Why this scoreMake target tools discoverable on the first callSelection accuracy 100%, destructive-action safety rate 100% (baseline only -- no rewrite pass applied).
Highest-impact fix
View full report
Clarify tool names, decision boundaries, and required argument schemas so an agent can choose and construct the target call without exploratory steps. |
||||||||
| 07 | 0xdaef0f/job-searchoor (job-searchoor) job-searchoor · v0.1.0 | 99.5/100 | A | 30.0/30 | 29.5/30 | 20.0/20 | 20.0/20 | |
Why this scoreRewrite 1 underspecified tool descriptionSelection accuracy 100→100, token cost up 7%, unconfirmed writes 100%→100%.
Highest-impact fix
View full report
State each tool's behavior, explain every parameter, and declare return semantics, prioritizing tools exercised by failed selection tasks. |
||||||||
| 08 | AntonioTF5/soul-mcp-server (soul-mcp-server) soul-mcp-server · v1.0.0 | 99.5/100 | A | 30.0/30 | 29.5/30 | 20.0/20 | 20.0/20 | |
Why this scoreRewrite 1 underspecified tool descriptionSelection accuracy 100→100, token cost up 3%, unconfirmed writes 100%→100%.
Highest-impact fix
View full report
State each tool's behavior, explain every parameter, and declare return semantics, prioritizing tools exercised by failed selection tasks. |
||||||||
| 09 | 930m310n/geomelon-mcp (geomelon-mcp) geomelon-mcp · v1.1.1 | 99.2/100 | A | 30.0/30 | 30.0/30 | 20.0/20 | 19.2/20 | |
Why this scoreMake target tools discoverable on the first callSelection accuracy 100%, destructive-action safety rate 100% (baseline only -- no rewrite pass applied).
Highest-impact fix
View full report
Clarify tool names, decision boundaries, and required argument schemas so an agent can choose and construct the target call without exploratory steps. |
||||||||
| 10 | stefan-xyz/mcp-server-runescape (mcp-server-runescape) mcp-server-runescape · v1.0.0 | 99.2/100 | A | 30.0/30 | 29.2/30 | 20.0/20 | 20.0/20 | |
Why this scoreDisambiguate overlapping tool choicesSelection accuracy 100%, destructive-action safety rate 100% (baseline only -- no rewrite pass applied).
Highest-impact fix
View full report
Give confusable tools explicit decision boundaries and examples of when to choose one instead of the other. |
||||||||
| 11 | aidevelopers2/remoteopenclaw-mcp (remoteopenclaw) remoteopenclaw · v0.2.0 | 99.0/100 | A | 30.0/30 | 29.0/30 | 20.0/20 | 20.0/20 | |
Why this scoreRewrite 3 underspecified tool descriptionsSelection accuracy 100→100, token cost up 5%, unconfirmed writes 100%→100%.
Highest-impact fix
View full report
State each tool's behavior, explain every parameter, and declare return semantics, prioritizing tools exercised by failed selection tasks. |
||||||||
| 12 | realwigu/mcp-doctor (@wigu/mcp-doctor) wigu-mcp-doctor · v0.3.0 | 98.9/100 | A | 30.0/30 | 29.3/30 | 20.0/20 | 19.7/20 | |
Why this scoreRewrite 3 underspecified tool descriptionsSelection accuracy 100→100, token cost up 7%, unconfirmed writes 100%→100%.
Highest-impact fix
View full report
State each tool's behavior, explain every parameter, and declare return semantics, prioritizing tools exercised by failed selection tasks. |
||||||||
| 13 | mobileshop9991-star/clipwise-mcp (clipwise-mcp-server) clipwise-mcp-server · v1.0.0 | 98.6/100 | A | 30.0/30 | 29.6/30 | 20.0/20 | 19.0/20 | |
Why this scoreMake target tools discoverable on the first callSelection accuracy 100%, destructive-action safety rate 100% (baseline only -- no rewrite pass applied).
Highest-impact fix
View full report
Clarify tool names, decision boundaries, and required argument schemas so an agent can choose and construct the target call without exploratory steps. |
||||||||
| 14 | alimuratkuslu/byok-observability-mcp (byok-observability-mcp) byok-observability-mcp · v0.1.0 | 98.6/100 | A | 30.0/30 | 29.5/30 | 20.0/20 | 19.1/20 | |
Why this scoreMake target tools discoverable on the first callSelection accuracy 100→100, token cost up 4%, unconfirmed writes 100%→100%.
Highest-impact fix
View full report
Clarify tool names, decision boundaries, and required argument schemas so an agent can choose and construct the target call without exploratory steps. |
||||||||
| 15 | gregm711/agent-domain-service-mcp (agent-domain-service-mcp) agent-domain-service-mcp · v1.0.0 | 98.5/100 | A | 30.0/30 | 29.1/30 | 20.0/20 | 19.4/20 | |
Why this scoreRewrite 1 underspecified tool descriptionSelection accuracy 100→100, token cost up 2%, unconfirmed writes 100%→100%.
Highest-impact fix
View full report
State each tool's behavior, explain every parameter, and declare return semantics, prioritizing tools exercised by failed selection tasks. |
||||||||
| 16 | fdcommercial/property-finance-mcp (@fdcommercial/property-finance-mcp) fdcommercial-property-finance-mcp · v0.1.0 | 98.5/100 | A | 30.0/30 | 30.0/30 | 19.3/20 | 19.2/20 | |
Why this scoreMake target tools discoverable on the first callSelection accuracy 100%, destructive-action safety rate 100% (baseline only -- no rewrite pass applied).
Highest-impact fix
View full report
Clarify tool names, decision boundaries, and required argument schemas so an agent can choose and construct the target call without exploratory steps. |
||||||||
| 17 | mambalabsdev/mcp-domain-to-linkedin-url-resolver (@mambalabsdev/mcp-domain-to-linkedin-url-resolver) mambalabsdev-mcp-domain-to-linkedin-url-resolver · v1.0.5 | 98.2/100 | A | 30.0/30 | 30.0/30 | 20.0/20 | 18.2/20 | |
Why this scoreMake target tools discoverable on the first callSelection accuracy 100%, destructive-action safety rate 100% (baseline only -- no rewrite pass applied).
Highest-impact fix
View full report
Clarify tool names, decision boundaries, and required argument schemas so an agent can choose and construct the target call without exploratory steps. |
||||||||
| 18 | muhannad-hash/git-context-mcp (git-context-mcp) git-context-mcp · v1.1.0 | 97.9/100 | A | 30.0/30 | 29.4/30 | 20.0/20 | 18.6/20 | |
Why this scoreMake target tools discoverable on the first callSelection accuracy 100→100, token cost up 2%, unconfirmed writes 100%→100%.
Highest-impact fix
View full report
Clarify tool names, decision boundaries, and required argument schemas so an agent can choose and construct the target call without exploratory steps. |
||||||||
| 19 | chefcohen/corroborate-mcp (corroborate-mcp) corroborate-mcp · v0.1.0 | 97.7/100 | A | 30.0/30 | 29.8/30 | 20.0/20 | 17.9/20 | |
Why this scoreMake target tools discoverable on the first callSelection accuracy 100→100, token cost up 3%, unconfirmed writes 100%→100%.
Highest-impact fix
View full report
Clarify tool names, decision boundaries, and required argument schemas so an agent can choose and construct the target call without exploratory steps. |
||||||||
| 20 | cantian-ai/bazi-mcp (bazi-mcp) bazi-mcp · v0.0.1 | 97.6/100 | A | 30.0/30 | 28.4/30 | 20.0/20 | 19.2/20 | |
Why this scoreRewrite 3 underspecified tool descriptionsSelection accuracy 98→100, token cost up 14%, unconfirmed writes 100%→100%.
Highest-impact fix
View full report
State each tool's behavior, explain every parameter, and declare return semantics, prioritizing tools exercised by failed selection tasks. |
||||||||
| 21 | mambalabsdev/mcp-company-firmographic-enricher (@mambalabsdev/mcp-company-firmographic-enricher) mambalabsdev-mcp-company-firmographic-enricher · v1.0.0 | 97.5/100 | A | 30.0/30 | 29.6/30 | 20.0/20 | 18.0/20 | |
Why this scoreMake target tools discoverable on the first callSelection accuracy 98%, destructive-action safety rate 100% (baseline only -- no rewrite pass applied).
Highest-impact fix
View full report
Clarify tool names, decision boundaries, and required argument schemas so an agent can choose and construct the target call without exploratory steps. |
||||||||
| 22 | qinisolabs/icdwise (icdwise) icdwise · v0.1.2 | 97.5/100 | A | 30.0/30 | 28.6/30 | 20.0/20 | 18.9/20 | |
Why this scoreDisambiguate overlapping tool choicesSelection accuracy 96%, destructive-action safety rate 100% (baseline only -- no rewrite pass applied).
Highest-impact fix
View full report
Give confusable tools explicit decision boundaries and examples of when to choose one instead of the other. |
||||||||
| 23 | Ihor-Sokoliuk/MCP-SearXNG (mcp-searxng) mcp-searxng · v1.14.0 | 97.0/100 | A | 30.0/30 | 29.5/30 | 19.9/20 | 17.7/20 | |
Why this scoreMake target tools discoverable on the first callSelection accuracy 100→100, token cost up 2%, unconfirmed writes 100%→100%.
Highest-impact fix
View full report
Clarify tool names, decision boundaries, and required argument schemas so an agent can choose and construct the target call without exploratory steps. |
||||||||
| 24 | chanmeng/google-news-mcp-server (@chanmeng666/google-news-server) chanmeng666-google-news-server · v1.0.0 | 96.9/100 | A | 30.0/30 | 29.8/30 | 20.0/20 | 17.1/20 | |
Why this scoreMake target tools discoverable on the first callSelection accuracy 100→100, token cost up 6%, unconfirmed writes 100%→100%.
Highest-impact fix
View full report
Clarify tool names, decision boundaries, and required argument schemas so an agent can choose and construct the target call without exploratory steps. |
||||||||
| 25 | AIweather-Anurag/ottasia-mcp-server (@ottasia/mcp-server) ottasia-mcp-server · v0.1.0 | 96.9/100 | A | 30.0/30 | 29.7/30 | 20.0/20 | 17.2/20 | |
Why this scoreMake target tools discoverable on the first callSelection accuracy 100%, destructive-action safety rate 100% (baseline only -- no rewrite pass applied).
Highest-impact fix
View full report
Clarify tool names, decision boundaries, and required argument schemas so an agent can choose and construct the target call without exploratory steps. |
||||||||
| 26 | muhannad-hash/mcp-shield (@muhannad-hash/mcp-shield) muhannad-hash-mcp-shield · v1.0.0 | 96.8/100 | A | 30.0/30 | 28.8/30 | 20.0/20 | 18.0/20 | |
Why this scoreMake target tools discoverable on the first callSelection accuracy 98→98, token cost up 3%, unconfirmed writes 100%→100%.
Highest-impact fix
View full report
Clarify tool names, decision boundaries, and required argument schemas so an agent can choose and construct the target call without exploratory steps. |
||||||||
| 27 | HasanJahidul/git-insight-mcp (git-insight-mcp) git-insight-mcp · v0.1.3 | 96.7/100 | A | 30.0/30 | 28.6/30 | 19.3/20 | 18.8/20 | |
Why this scoreDisambiguate overlapping tool choicesSelection accuracy 100%, destructive-action safety rate 100% (baseline only -- no rewrite pass applied).
Highest-impact fix
View full report
Give confusable tools explicit decision boundaries and examples of when to choose one instead of the other. |
||||||||
| 28 | thevibepreneur/gapbase-mcp (gapbase-mcp) gapbase-mcp · v0.1.0 | 96.7/100 | A | 30.0/30 | 29.6/30 | 20.0/20 | 17.1/20 | |
Why this scoreMake target tools discoverable on the first callSelection accuracy 100%, destructive-action safety rate 100% (baseline only -- no rewrite pass applied).
Highest-impact fix
View full report
Clarify tool names, decision boundaries, and required argument schemas so an agent can choose and construct the target call without exploratory steps. |
||||||||
| 29 | shuji-bonji/xcomet-mcp-server (xcomet-mcp-server) xcomet-mcp-server · v0.6.3 | 96.4/100 | A | 30.0/30 | 28.3/30 | 19.0/20 | 19.2/20 | |
Why this scoreDisambiguate overlapping tool choicesSelection accuracy 96%, destructive-action safety rate 100% (baseline only -- no rewrite pass applied).
Highest-impact fix
View full report
Give confusable tools explicit decision boundaries and examples of when to choose one instead of the other. |
||||||||
| 30 | gregario/onepiece-oracle (onepiece-oracle) onepiece-oracle · v0.1.2 | 96.3/100 | A | 30.0/30 | 27.9/30 | 20.0/20 | 18.4/20 | |
Why this scoreRewrite 5 underspecified tool descriptionsSelection accuracy 98→98, token cost up 4%, unconfirmed writes 100%→100%.
Highest-impact fix
View full report
State each tool's behavior, explain every parameter, and declare return semantics, prioritizing tools exercised by failed selection tasks. |
||||||||
| 31 | BluesPrince/thiri-mcp (@bluesprincemedia/thiri-mcp) bluesprincemedia-thiri-mcp · v0.3.0 | 96.0/100 | A | 30.0/30 | 29.9/30 | 20.0/20 | 16.2/20 | |
Why this scoreMake target tools discoverable on the first callSelection accuracy 100%, destructive-action safety rate 100% (baseline only -- no rewrite pass applied).
Highest-impact fix
View full report
Clarify tool names, decision boundaries, and required argument schemas so an agent can choose and construct the target call without exploratory steps. |
||||||||
| 32 | karanb192/reddit-mcp-buddy (reddit-mcp-buddy) reddit-mcp-buddy · v1.1.14 | 95.9/100 | A | 30.0/30 | 28.4/30 | 19.8/20 | 17.7/20 | |
Why this scoreMake target tools discoverable on the first callSelection accuracy 96→96, token cost up 6%, unconfirmed writes 100%→100%.
Highest-impact fix
View full report
Clarify tool names, decision boundaries, and required argument schemas so an agent can choose and construct the target call without exploratory steps. |
||||||||
| 33 | ofershap/mcp-server-markdown (mcp-server-markdown) mcp-server-markdown · v1.0.0 | 95.8/100 | A | 30.0/30 | 28.7/30 | 20.0/20 | 17.1/20 | |
Why this scoreMake target tools discoverable on the first callSelection accuracy 100→100, token cost up 4%, unconfirmed writes 100%→100%.
Highest-impact fix
View full report
Clarify tool names, decision boundaries, and required argument schemas so an agent can choose and construct the target call without exploratory steps. |
||||||||
| 34 | michalperni11-gif/secfinapi-mcp (secfinapi-mcp) secfinapi-mcp · v0.1.0 | 95.5/100 | A | 30.0/30 | 29.1/30 | 20.0/20 | 16.4/20 | |
Why this scoreMake target tools discoverable on the first callSelection accuracy 100%, destructive-action safety rate 100% (baseline only -- no rewrite pass applied).
Highest-impact fix
View full report
Clarify tool names, decision boundaries, and required argument schemas so an agent can choose and construct the target call without exploratory steps. |
||||||||
| 35 | sthan-io/mcp-server (@sthan/mcp-server) sthan-mcp-server · v0.1.5 | 95.4/100 | A | 30.0/30 | 28.9/30 | 19.7/20 | 16.9/20 | |
Why this scoreMake target tools discoverable on the first callSelection accuracy 100%, destructive-action safety rate 100% (baseline only -- no rewrite pass applied).
Highest-impact fix
View full report
Clarify tool names, decision boundaries, and required argument schemas so an agent can choose and construct the target call without exploratory steps. |
||||||||
| 36 | tunedforai/x402-mcp (@tunedforai/x402-mcp) tunedforai-x402-mcp · v1.0.0 | 95.4/100 | A | 30.0/30 | 25.8/30 | 20.0/20 | 19.6/20 | |
Why this scoreRewrite 6 underspecified tool descriptionsSelection accuracy 98→100, token cost up 4%, unconfirmed writes 100%→100%.
Highest-impact fix
View full report
State each tool's behavior, explain every parameter, and declare return semantics, prioritizing tools exercised by failed selection tasks. |
||||||||
| 37 | shieldly-io/mcp (@shieldly/mcp) shieldly-mcp · v1.0.1 | 95.0/100 | A | 30.0/30 | 27.2/30 | 20.0/20 | 17.8/20 | |
Why this scoreRewrite 2 underspecified tool descriptionsSelection accuracy 89→97, token cost up 2%, unconfirmed writes 100%→100%.
Highest-impact fix
View full report
State each tool's behavior, explain every parameter, and declare return semantics, prioritizing tools exercised by failed selection tasks. |
||||||||
| 38 | agentbodegastore/agentbodega (@agentbodega/mcp) agentbodega-mcp · v0.1.7 | 94.7/100 | A | 30.0/30 | 28.4/30 | 20.0/20 | 16.2/20 | |
Why this scoreMake target tools discoverable on the first callSelection accuracy 100→100, token cost up 1%, unconfirmed writes 100%→100%.
Highest-impact fix
View full report
Clarify tool names, decision boundaries, and required argument schemas so an agent can choose and construct the target call without exploratory steps. |
||||||||
| 39 | mlava/scholar-sidekick-mcp (scholar-sidekick-mcp) scholar-sidekick-mcp · v0.8.7 | 94.5/100 | A | 30.0/30 | 28.7/30 | 17.1/20 | 18.8/20 | |
Why this scoreReduce the tool-definition and task token payloadSelection accuracy 98→98, token cost up 2%, unconfirmed writes 100%→100%.
Highest-impact fix
View full report
Consolidate overlapping tools and trim repetitive schema text while preserving behavior, parameter, and return semantics. |
||||||||
| 40 | perceptdot/percept (@perceptdot/core) perceptdot-core · v0.1.0 | 94.3/100 | A | 30.0/30 | 28.6/30 | 20.0/20 | 15.7/20 | |
Why this scoreMake target tools discoverable on the first callSelection accuracy 100→100, token cost up 4%, unconfirmed writes 100%→100%.
Highest-impact fix
View full report
Clarify tool names, decision boundaries, and required argument schemas so an agent can choose and construct the target call without exploratory steps. |
||||||||
| 41 | tradallo/reputation (@tradallo/reputation) tradallo-reputation · v0.3.2 | 94.3/100 | A | 30.0/30 | 28.8/30 | 19.9/20 | 15.6/20 | |
Why this scoreMake target tools discoverable on the first callSelection accuracy 98→95, token cost down 0%, unconfirmed writes 100%→100%.
Highest-impact fix
View full report
Clarify tool names, decision boundaries, and required argument schemas so an agent can choose and construct the target call without exploratory steps. |
||||||||
| 42 | shunshi-ai/bazi-reader-mcp (shunshi-bazi-mcp) shunshi-bazi-mcp · v0.1.0 | 94.2/100 | A | 30.0/30 | 29.4/30 | 19.8/20 | 15.0/20 | |
Why this scoreMake target tools discoverable on the first callSelection accuracy 98%, destructive-action safety rate 100% (baseline only -- no rewrite pass applied).
Highest-impact fix
View full report
Clarify tool names, decision boundaries, and required argument schemas so an agent can choose and construct the target call without exploratory steps. |
||||||||
| 43 | Darko893/haunt-mcp-server (@hauntapi/mcp-server) hauntapi-mcp-server · v1.1.2 | 94.0/100 | A | 30.0/30 | 27.8/30 | 19.8/20 | 16.5/20 | |
Why this scoreMake target tools discoverable on the first callSelection accuracy 98→95, token cost down 0%, unconfirmed writes 100%→100%.
Highest-impact fix
View full report
Clarify tool names, decision boundaries, and required argument schemas so an agent can choose and construct the target call without exploratory steps. |
||||||||
| 44 | ofershap/mcp-server-dns (mcp-server-dns) mcp-server-dns · v1.0.0 | 93.8/100 | A | 30.0/30 | 28.0/30 | 20.0/20 | 15.7/20 | |
Why this scoreMake target tools discoverable on the first callSelection accuracy 98→98, token cost up 4%, unconfirmed writes 100%→100%.
Highest-impact fix
View full report
Clarify tool names, decision boundaries, and required argument schemas so an agent can choose and construct the target call without exploratory steps. |
||||||||
| 45 | ofershap/mcp-server-npm-plus (mcp-server-npm-plus) mcp-server-npm-plus · v1.0.0 | 93.7/100 | A | 30.0/30 | 28.4/30 | 20.0/20 | 15.3/20 | |
Why this scoreMake target tools discoverable on the first callSelection accuracy 100→100, token cost up 6%, unconfirmed writes 100%→100%.
Highest-impact fix
View full report
Clarify tool names, decision boundaries, and required argument schemas so an agent can choose and construct the target call without exploratory steps. |
||||||||
| 46 | ejwhite7/brandkit-mcp (brandkit-mcp) brandkit-mcp · v2.1.0 | 93.6/100 | A | 30.0/30 | 24.0/30 | 19.9/20 | 19.7/20 | |
Why this scoreRewrite 10 underspecified tool descriptionsSelection accuracy 100→100, token cost up 13%, unconfirmed writes 100%→100%.
Highest-impact fix
View full report
State each tool's behavior, explain every parameter, and declare return semantics, prioritizing tools exercised by failed selection tasks. |
||||||||
| 47 | QuantGeekDev/coincap-mcp (coincap-mcp) coincap-mcp · v0.9.2 | 93.6/100 | A | 30.0/30 | 28.7/30 | 20.0/20 | 14.9/20 | |
Why this scoreMake target tools discoverable on the first callSelection accuracy 100→100, token cost up 7%, unconfirmed writes 100%→100%.
Highest-impact fix
View full report
Clarify tool names, decision boundaries, and required argument schemas so an agent can choose and construct the target call without exploratory steps. |
||||||||
| 48 | inkog-io/inkog-mcp (@inkog-io/mcp) inkog-io-mcp · v1.0.0 | 93.5/100 | A | 30.0/30 | 24.6/30 | 19.7/20 | 19.3/20 | |
Why this scoreRewrite 10 underspecified tool descriptionsSelection accuracy 98→100, token cost up 12%, unconfirmed writes 100%→100%.
Highest-impact fix
View full report
State each tool's behavior, explain every parameter, and declare return semantics, prioritizing tools exercised by failed selection tasks. |
||||||||
| 49 | robhunter/agentdeals (agentdeals) agentdeals · v0.1.0 | 93.5/100 | A | 30.0/30 | 26.1/30 | 19.4/20 | 17.9/20 | |
Why this scoreRewrite 8 underspecified tool descriptionsSelection accuracy 100→100, token cost down 0%, unconfirmed writes 100%→100%.
Highest-impact fix
View full report
State each tool's behavior, explain every parameter, and declare return semantics, prioritizing tools exercised by failed selection tasks. |
||||||||
| 50 | Context7 (Upstash, live library docs lookup) context7-mcp · v3.2.5 | 93.5/100 | A | 30.0/30 | 30.0/30 | 20.0/20 | 13.4/20 | |
Why this scoreMake target tools discoverable on the first callSelection accuracy 100%, destructive-action safety rate 100% (baseline only -- no rewrite pass applied).
Highest-impact fix
View full report
Clarify tool names, decision boundaries, and required argument schemas so an agent can choose and construct the target call without exploratory steps. |
||||||||
| 51 | forgemeshlabs/travel-mcp (@forgemeshlabs/travel-assistant-mcp) forgemeshlabs-travel-assistant-mcp · v0.1.2 | 93.4/100 | A | 30.0/30 | 30.0/30 | 20.0/20 | 13.4/20 | |
Why this scoreMake target tools discoverable on the first callSelection accuracy 100%, destructive-action safety rate 100% (baseline only -- no rewrite pass applied).
Highest-impact fix
View full report
Clarify tool names, decision boundaries, and required argument schemas so an agent can choose and construct the target call without exploratory steps. |
||||||||
| 52 | gregario/astronomy-oracle (astronomy-oracle) astronomy-oracle · v0.1.0 | 93.3/100 | A | 30.0/30 | 29.5/30 | 20.0/20 | 13.9/20 | |
Why this scoreMake target tools discoverable on the first callSelection accuracy 100→100, token cost up 2%, unconfirmed writes 100%→100%.
Highest-impact fix
View full report
Clarify tool names, decision boundaries, and required argument schemas so an agent can choose and construct the target call without exploratory steps. |
||||||||
| 53 | louis030195/easy-obsidian-mcp (@louis030195/mcp-obsidian) louis030195-mcp-obsidian · v0.1.0 | 93.3/100 | A | 30.0/30 | 26.1/30 | 19.8/20 | 17.5/20 | |
Why this scoreRewrite 5 underspecified tool descriptionsSelection accuracy 95→98, token cost up 7%, unconfirmed writes 100%→100%.
Highest-impact fix
View full report
State each tool's behavior, explain every parameter, and declare return semantics, prioritizing tools exercised by failed selection tasks. |
||||||||
| 54 | jamesdfinance-dev/lazaretto-mcp (lazaretto-mcp) lazaretto-mcp · v0.1.0 | 93.3/100 | A | 30.0/30 | 29.3/30 | 20.0/20 | 14.0/20 | |
Why this scoreMake target tools discoverable on the first callSelection accuracy 100→100, token cost up 11%, unconfirmed writes 100%→100%.
Highest-impact fix
View full report
Clarify tool names, decision boundaries, and required argument schemas so an agent can choose and construct the target call without exploratory steps. |
||||||||
| 55 | AgentBase1/mcp-server (agentbase-mcp) agentbase-mcp · v1.0.0 | 93.1/100 | A | 30.0/30 | 29.7/30 | 20.0/20 | 13.4/20 | |
Why this scoreMake target tools discoverable on the first callSelection accuracy 100%, destructive-action safety rate 100% (baseline only -- no rewrite pass applied).
Highest-impact fix
View full report
Clarify tool names, decision boundaries, and required argument schemas so an agent can choose and construct the target call without exploratory steps. |
||||||||
| 56 | Bishop81/wpgoldmine-mcp (wpgoldmine-mcp) wpgoldmine-mcp · v0.1.0 | 92.9/100 | A | 30.0/30 | 29.5/30 | 20.0/20 | 13.4/20 | |
Why this scoreMake target tools discoverable on the first callSelection accuracy 100→100, token cost up 1%, unconfirmed writes 100%→100%.
Highest-impact fix
View full report
Clarify tool names, decision boundaries, and required argument schemas so an agent can choose and construct the target call without exploratory steps. |
||||||||
| 57 | Daichi-Kudo/llm-advisor-mcp (llm-advisor-mcp) llm-advisor-mcp · v0.4.0 | 92.8/100 | A | 30.0/30 | 29.2/30 | 20.0/20 | 13.6/20 | |
Why this scoreMake target tools discoverable on the first callSelection accuracy 100%, destructive-action safety rate 100% (baseline only -- no rewrite pass applied).
Highest-impact fix
View full report
Clarify tool names, decision boundaries, and required argument schemas so an agent can choose and construct the target call without exploratory steps. |
||||||||
| 58 | mambalabsdev/mcp-funding-press-signal-scanner (@mambalabsdev/mcp-funding-press-signal-scanner) mambalabsdev-mcp-funding-press-signal-scanner · v1.0.0 | 92.8/100 | A | 30.0/30 | 30.0/30 | 20.0/20 | 12.8/20 | |
Why this scoreMake target tools discoverable on the first callSelection accuracy 100%, destructive-action safety rate 100% (baseline only -- no rewrite pass applied).
Highest-impact fix
View full report
Clarify tool names, decision boundaries, and required argument schemas so an agent can choose and construct the target call without exploratory steps. |
||||||||
| 59 | calllint/calllint (calllint-mcp) calllint-mcp · v0.2.0 | 92.7/100 | A | 30.0/30 | 28.4/30 | 20.0/20 | 14.3/20 | |
Why this scoreMake target tools discoverable on the first callSelection accuracy 93%, destructive-action safety rate 100% (baseline only -- no rewrite pass applied).
Highest-impact fix
View full report
Clarify tool names, decision boundaries, and required argument schemas so an agent can choose and construct the target call without exploratory steps. |
||||||||
| 60 | ScrapeUnblocker/scrapeunblocker-mcp (scrapeunblocker-mcp) scrapeunblocker-mcp · v0.1.2 | 92.5/100 | A | 30.0/30 | 29.7/30 | 20.0/20 | 12.8/20 | |
Why this scoreMake target tools discoverable on the first callSelection accuracy 100%, destructive-action safety rate 100% (baseline only -- no rewrite pass applied).
Highest-impact fix
View full report
Clarify tool names, decision boundaries, and required argument schemas so an agent can choose and construct the target call without exploratory steps. |
||||||||
| 61 | gregario/brewers-almanack (brewers-almanack) brewers-almanack · v0.2.7 | 92.4/100 | A | 30.0/30 | 28.0/30 | 19.4/20 | 15.0/20 | |
Why this scoreMake target tools discoverable on the first callSelection accuracy 100→100, token cost up 2%, unconfirmed writes 100%→100%.
Highest-impact fix
View full report
Clarify tool names, decision boundaries, and required argument schemas so an agent can choose and construct the target call without exploratory steps. |
||||||||
| 62 | snenenenenenene/transita-mcp-server (@transita/mcp-server) transita-mcp-server · v0.1.3 | 92.4/100 | A | 30.0/30 | 27.4/30 | 20.0/20 | 15.0/20 | |
Why this scoreMake target tools discoverable on the first callSelection accuracy 96→100, token cost up 8%, unconfirmed writes 100%→100%.
Highest-impact fix
View full report
Clarify tool names, decision boundaries, and required argument schemas so an agent can choose and construct the target call without exploratory steps. |
||||||||
| 63 | beshogun/planwire-mcp (planwire-mcp) planwire-mcp · v0.1.4 | 92.4/100 | A | 30.0/30 | 28.3/30 | 20.0/20 | 14.1/20 | |
Why this scoreMake target tools discoverable on the first callSelection accuracy 98→100, token cost up 5%, unconfirmed writes 100%→100%.
Highest-impact fix
View full report
Clarify tool names, decision boundaries, and required argument schemas so an agent can choose and construct the target call without exploratory steps. |
||||||||
| 64 | Backspace-me/sportscore-mcp (sportscore-mcp) sportscore-mcp · v0.1.0 | 92.3/100 | A | 30.0/30 | 27.7/30 | 20.0/20 | 14.7/20 | |
Why this scoreMake target tools discoverable on the first callSelection accuracy 100→100, token cost up 6%, unconfirmed writes 100%→100%.
Highest-impact fix
View full report
Clarify tool names, decision boundaries, and required argument schemas so an agent can choose and construct the target call without exploratory steps. |
||||||||
| 65 | arXiv MCP Server (@cyanheads/arxiv-mcp-server, Smithery popular) cyanheads-arxiv-mcp-server · v0.1.0 | 92.3/100 | A | 30.0/30 | 29.1/30 | 18.3/20 | 14.8/20 | |
Why this scoreMake target tools discoverable on the first callSelection accuracy 98%, destructive-action safety rate 100% (baseline only -- no rewrite pass applied).
Highest-impact fix
View full report
Clarify tool names, decision boundaries, and required argument schemas so an agent can choose and construct the target call without exploratory steps. |
||||||||
| 66 | MobileReality/mdma (@mobile-reality/mdma-mcp) mobile-reality-mdma-mcp · v0.2.4 | 92.2/100 | A | 30.0/30 | 28.2/30 | 19.9/20 | 14.1/20 | |
Why this scoreMake target tools discoverable on the first callSelection accuracy 98%, destructive-action safety rate 100% (baseline only -- no rewrite pass applied).
Highest-impact fix
View full report
Clarify tool names, decision boundaries, and required argument schemas so an agent can choose and construct the target call without exploratory steps. |
||||||||
| 67 | ofershap/mcp-server-scraper (mcp-server-scraper) mcp-server-scraper · v1.0.0 | 92.2/100 | A | 30.0/30 | 28.7/30 | 20.0/20 | 13.5/20 | |
Why this scoreMake target tools discoverable on the first callSelection accuracy 98→100, token cost up 3%, unconfirmed writes 100%→100%.
Highest-impact fix
View full report
Clarify tool names, decision boundaries, and required argument schemas so an agent can choose and construct the target call without exploratory steps. |
||||||||
| 68 | KyrieTangSheng/mcp-server-nationalparks (mcp-server-nationalparks) mcp-server-nationalparks · v1.0.0 | 92.1/100 | A | 30.0/30 | 26.6/30 | 20.0/20 | 15.5/20 | |
Why this scoreMake target tools discoverable on the first callSelection accuracy 100→100, token cost up 12%, unconfirmed writes 100%→100%.
Highest-impact fix
View full report
Clarify tool names, decision boundaries, and required argument schemas so an agent can choose and construct the target call without exploratory steps. |
||||||||
| 69 | haomingkoo/japan-seasons-mcp (japan-seasons-mcp) japan-seasons-mcp · v0.5.0 | 92.1/100 | A | 30.0/30 | 27.2/30 | 17.0/20 | 17.9/20 | |
Why this scoreReduce the tool-definition and task token payloadSelection accuracy 93%, destructive-action safety rate 100% (baseline only -- no rewrite pass applied).
Highest-impact fix
View full report
Consolidate overlapping tools and trim repetitive schema text while preserving behavior, parameter, and return semantics. |
||||||||
| 70 | gpu-bridge/mcp-server (@gpu-bridge/mcp-server) gpu-bridge-mcp-server · v2.0.0 | 92.1/100 | A | 30.0/30 | 28.1/30 | 20.0/20 | 14.0/20 | |
Why this scoreMake target tools discoverable on the first callSelection accuracy 100→98, token cost up 4%, unconfirmed writes 100%→100%.
Highest-impact fix
View full report
Clarify tool names, decision boundaries, and required argument schemas so an agent can choose and construct the target call without exploratory steps. |
||||||||
| 71 | woladi/macos-vision-mcp (macos-vision-mcp) macos-vision-mcp · v0.4.9 | 92.0/100 | A | 30.0/30 | 27.8/30 | 19.6/20 | 14.6/20 | |
Why this scoreMake target tools discoverable on the first callSelection accuracy 93%, destructive-action safety rate 100% (baseline only -- no rewrite pass applied).
Highest-impact fix
View full report
Clarify tool names, decision boundaries, and required argument schemas so an agent can choose and construct the target call without exploratory steps. |
||||||||
| 72 | iplocate/mcp-server-iplocate (@iplocate/mcp-server) iplocate-mcp-server · v1.0.0 | 91.9/100 | A | 30.0/30 | 28.9/30 | 20.0/20 | 13.0/20 | |
Why this scoreMake target tools discoverable on the first callSelection accuracy 100%, destructive-action safety rate 100% (baseline only -- no rewrite pass applied).
Highest-impact fix
View full report
Clarify tool names, decision boundaries, and required argument schemas so an agent can choose and construct the target call without exploratory steps. |
||||||||
| 73 | gregario/warhammer-oracle (warhammer-oracle) warhammer-oracle · v0.2.42 | 91.9/100 | A | 30.0/30 | 26.8/30 | 18.1/20 | 16.9/20 | |
Why this scoreRewrite 5 underspecified tool descriptionsSelection accuracy 98→95, token cost up 1%, unconfirmed writes 100%→100%.
Highest-impact fix
View full report
State each tool's behavior, explain every parameter, and declare return semantics, prioritizing tools exercised by failed selection tasks. |
||||||||
| 74 | Bigsy/Clojars-MCP-Server (clojars-deps-server) clojars-deps-server · v0.2.0 | 91.9/100 | A | 30.0/30 | 26.8/30 | 20.0/20 | 15.1/20 | |
Why this scoreMake target tools discoverable on the first callSelection accuracy 95→98, token cost up 12%, unconfirmed writes 100%→100%.
Highest-impact fix
View full report
Clarify tool names, decision boundaries, and required argument schemas so an agent can choose and construct the target call without exploratory steps. |
||||||||
| 75 | tadas-github/a2asearch-mcp (a2asearch-mcp) a2asearch-mcp · v1.0.0 | 91.7/100 | A | 30.0/30 | 28.9/30 | 20.0/20 | 12.8/20 | |
Why this scoreMake target tools discoverable on the first callSelection accuracy 100→100, token cost up 5%, unconfirmed writes 100%→100%.
Highest-impact fix
View full report
Clarify tool names, decision boundaries, and required argument schemas so an agent can choose and construct the target call without exploratory steps. |
||||||||
| 76 | cqtrinv/trinvmcp (mcp-trinv-server) mcp-trinv-server · v0.1.1 | 91.6/100 | A | 30.0/30 | 29.4/30 | 20.0/20 | 12.2/20 | |
Why this scoreMake target tools discoverable on the first callSelection accuracy 100→98, token cost up 1%, unconfirmed writes 100%→100%.
Highest-impact fix
View full report
Clarify tool names, decision boundaries, and required argument schemas so an agent can choose and construct the target call without exploratory steps. |
||||||||
| 77 | tresor4k/macalc-mcp (macalc-mcp) macalc-mcp · v0.4.0 | 91.5/100 | A | 30.0/30 | 27.5/30 | 18.6/20 | 15.4/20 | |
Why this scoreMake target tools discoverable on the first callSelection accuracy 100→98, token cost up 7%, unconfirmed writes 100%→100%.
Highest-impact fix
View full report
Clarify tool names, decision boundaries, and required argument schemas so an agent can choose and construct the target call without exploratory steps. |
||||||||
| 78 | MattJeff/orizn-mcp-server (orizn-visa-mcp) orizn-visa-mcp · v1.0.0 | 91.4/100 | A | 30.0/30 | 26.4/30 | 20.0/20 | 15.0/20 | |
Why this scoreMake target tools discoverable on the first callSelection accuracy 83%, destructive-action safety rate 100% (baseline only -- no rewrite pass applied).
Highest-impact fix
View full report
Clarify tool names, decision boundaries, and required argument schemas so an agent can choose and construct the target call without exploratory steps. |
||||||||
| 79 | doggybee/mcp-server-leetcode (@mcpfun/mcp-server-leetcode) mcpfun-mcp-server-leetcode · v1.0.0 | 91.4/100 | A | 30.0/30 | 26.9/30 | 20.0/20 | 14.5/20 | |
Why this scoreMake target tools discoverable on the first callSelection accuracy 100→100, token cost up 12%, unconfirmed writes 100%→100%.
Highest-impact fix
View full report
Clarify tool names, decision boundaries, and required argument schemas so an agent can choose and construct the target call without exploratory steps. |
||||||||
| 80 | pranciskus/newsmcp (@newsmcp/server) newsmcp-server · v1.0.2 | 91.3/100 | A | 30.0/30 | 29.6/30 | 20.0/20 | 11.7/20 | |
Why this scoreMake target tools discoverable on the first callSelection accuracy 100%, destructive-action safety rate 100% (baseline only -- no rewrite pass applied).
Highest-impact fix
View full report
Clarify tool names, decision boundaries, and required argument schemas so an agent can choose and construct the target call without exploratory steps. |
||||||||
| 81 | evanatpizzarobot/vr-org-mcp (vr-org-mcp) vr-org-mcp · v0.3.2 | 91.2/100 | A | 30.0/30 | 28.4/30 | 20.0/20 | 12.8/20 | |
Why this scoreMake target tools discoverable on the first callSelection accuracy 98→98, token cost up 4%, unconfirmed writes 100%→100%.
Highest-impact fix
View full report
Clarify tool names, decision boundaries, and required argument schemas so an agent can choose and construct the target call without exploratory steps. |
||||||||
| 82 | conorbronsdon/op3-mcp (@conorbronsdon/op3-mcp) conorbronsdon-op3-mcp · v0.2.0 | 91.1/100 | A | 30.0/30 | 28.5/30 | 19.8/20 | 12.8/20 | |
Why this scoreMake target tools discoverable on the first callSelection accuracy 100%, destructive-action safety rate 100% (baseline only -- no rewrite pass applied).
Highest-impact fix
View full report
Clarify tool names, decision boundaries, and required argument schemas so an agent can choose and construct the target call without exploratory steps. |
||||||||
| 83 | kehvinbehvin/json-mcp-filter (json-mcp-filter) json-mcp-filter · v1.0.1 | 91.1/100 | A | 30.0/30 | 29.5/30 | 20.0/20 | 11.6/20 | |
Why this scoreMake target tools discoverable on the first callSelection accuracy 100%, destructive-action safety rate 100% (baseline only -- no rewrite pass applied).
Highest-impact fix
View full report
Clarify tool names, decision boundaries, and required argument schemas so an agent can choose and construct the target call without exploratory steps. |
||||||||
| 84 | exa-labs/exa-mcp-server (exa-mcp-server) exa-mcp-server · v3.4.0 | 91.0/100 | A | 30.0/30 | 29.1/30 | 20.0/20 | 12.0/20 | |
Why this scoreMake target tools discoverable on the first callSelection accuracy 95%, destructive-action safety rate 100% (baseline only -- no rewrite pass applied).
Highest-impact fix
View full report
Clarify tool names, decision boundaries, and required argument schemas so an agent can choose and construct the target call without exploratory steps. |
||||||||
| 85 | uchit/mcp-regulated-ai-compliance (@hellouchit/mcp-regulated-ai-compliance) hellouchit-mcp-regulated-ai-compliance · v0.2.1 | 91.0/100 | A | 30.0/30 | 29.8/30 | 19.4/20 | 11.9/20 | |
Why this scoreMake target tools discoverable on the first callSelection accuracy 100→100, token cost up 1%, unconfirmed writes 100%→100%.
Highest-impact fix
View full report
Clarify tool names, decision boundaries, and required argument schemas so an agent can choose and construct the target call without exploratory steps. |
||||||||
| 86 | ssatama/rescuedogs-mcp-server (rescuedogs-mcp-server) rescuedogs-mcp-server · v1.2.0 | 91.0/100 | A | 30.0/30 | 27.1/30 | 19.2/20 | 14.7/20 | |
Why this scoreMake target tools discoverable on the first callSelection accuracy 98→98, token cost up 6%, unconfirmed writes 100%→100%.
Highest-impact fix
View full report
Clarify tool names, decision boundaries, and required argument schemas so an agent can choose and construct the target call without exploratory steps. |
||||||||
| 87 | romanshumy/llm-prices-data (@modelpricewatch/mcp) modelpricewatch-mcp · v1.0.0 | 90.6/100 | A | 30.0/30 | 28.7/30 | 20.0/20 | 11.9/20 | |
Why this scoreMake target tools discoverable on the first callSelection accuracy 96%, destructive-action safety rate 100% (baseline only -- no rewrite pass applied).
Highest-impact fix
View full report
Clarify tool names, decision boundaries, and required argument schemas so an agent can choose and construct the target call without exploratory steps. |
||||||||
| 88 | echojobsio/jdl-mcp-server (@jobdatalake/mcp-server) jobdatalake-mcp-server · v1.0.0 | 90.5/100 | A | 30.0/30 | 28.0/30 | 20.0/20 | 12.6/20 | |
Why this scoreMake target tools discoverable on the first callSelection accuracy 100→98, token cost up 10%, unconfirmed writes 100%→100%.
Highest-impact fix
View full report
Clarify tool names, decision boundaries, and required argument schemas so an agent can choose and construct the target call without exploratory steps. |
||||||||
| 89 | bamwor-dev/bamwor-mcp-server (bamwor-mcp-server) bamwor-mcp-server · v0.2.2 | 90.4/100 | A | 30.0/30 | 28.0/30 | 19.8/20 | 12.6/20 | |
Why this scoreMake target tools discoverable on the first callSelection accuracy 98→100, token cost up 2%, unconfirmed writes 100%→100%.
Highest-impact fix
View full report
Clarify tool names, decision boundaries, and required argument schemas so an agent can choose and construct the target call without exploratory steps. |
||||||||
| 90 | entire-vc/evc-spark-mcp (evc-spark-mcp) evc-spark-mcp · v1.0.0 | 90.1/100 | A | 30.0/30 | 27.6/30 | 20.0/20 | 12.5/20 | |
Why this scoreMake target tools discoverable on the first callSelection accuracy 95→95, token cost up 3%, unconfirmed writes 100%→100%.
Highest-impact fix
View full report
Clarify tool names, decision boundaries, and required argument schemas so an agent can choose and construct the target call without exploratory steps. |
||||||||
| 91 | mcpware/ui-annotator-mcp (@mcpware/ui-annotator) mcpware-ui-annotator · v0.1.0 | 90.1/100 | A | 30.0/30 | 28.0/30 | 20.0/20 | 12.1/20 | |
Why this scoreMake target tools discoverable on the first callSelection accuracy 95→98, token cost up 6%, unconfirmed writes 100%→100%.
Highest-impact fix
View full report
Clarify tool names, decision boundaries, and required argument schemas so an agent can choose and construct the target call without exploratory steps. |
||||||||
| 92 | openfate-ai/bazi-mcp (@openfate/bazi-mcp) openfate-bazi-mcp · v0.2.6 | 90.0/100 | B | 30.0/30 | 28.3/30 | 19.5/20 | 12.2/20 | |
Why this scoreMake target tools discoverable on the first callSelection accuracy 100→100, token cost up 5%, unconfirmed writes 100%→100%.
Highest-impact fix
View full report
Clarify tool names, decision boundaries, and required argument schemas so an agent can choose and construct the target call without exploratory steps. |
||||||||
| 93 | izzzzzi/codewiki-mcp (codewiki-mcp) codewiki-mcp · v1.1.2 | 89.5/100 | B | 30.0/30 | 26.6/30 | 20.0/20 | 12.9/20 | |
Why this scoreMake target tools discoverable on the first callSelection accuracy 98→100, token cost up 20%, unconfirmed writes 100%→100%.
Highest-impact fix
View full report
Clarify tool names, decision boundaries, and required argument schemas so an agent can choose and construct the target call without exploratory steps. |
||||||||
| 94 | OilpriceAPI/mcp-server (oilpriceapi-mcp) oilpriceapi-mcp · v3.0.0 | 89.5/100 | B | 30.0/30 | 25.8/30 | 17.4/20 | 16.3/20 | |
Why this scoreRewrite 8 underspecified tool descriptionsSelection accuracy 100→100, token cost up 2%, unconfirmed writes 100%→100%.
Highest-impact fix
View full report
State each tool's behavior, explain every parameter, and declare return semantics, prioritizing tools exercised by failed selection tasks. |
||||||||
| 95 | comparedge/mcp-server-comparedge (@comparedge/mcp-server) comparedge-mcp-server · v2.7.1 | 88.8/100 | B | 30.0/30 | 27.6/30 | 19.4/20 | 11.8/20 | |
Why this scoreMake target tools discoverable on the first callSelection accuracy 95%, destructive-action safety rate 100% (baseline only -- no rewrite pass applied).
Highest-impact fix
View full report
Clarify tool names, decision boundaries, and required argument schemas so an agent can choose and construct the target call without exploratory steps. |
||||||||
| 96 | jordanlyall/wc26-mcp (wc26-mcp) wc26-mcp · v0.3.1 | 88.7/100 | B | 30.0/30 | 26.3/30 | 18.5/20 | 13.8/20 | |
Why this scoreMake target tools discoverable on the first callSelection accuracy 98→98, token cost down 2%, unconfirmed writes 100%→100%.
Highest-impact fix
View full report
Clarify tool names, decision boundaries, and required argument schemas so an agent can choose and construct the target call without exploratory steps. |
||||||||
| 97 | serkan-ozal/driflyte-mcp-server (@driflyte/mcp-server) driflyte-mcp-server · v0.1.15 | 88.6/100 | B | 30.0/30 | 30.0/30 | 20.0/20 | 8.6/20 | |
Why this scoreMake target tools discoverable on the first callSelection accuracy 100%, destructive-action safety rate 100% (baseline only -- no rewrite pass applied).
Highest-impact fix
View full report
Clarify tool names, decision boundaries, and required argument schemas so an agent can choose and construct the target call without exploratory steps. |
||||||||
| 98 | APOGEOAPI/apogeoapi-mcp (@apogeoapi/mcp) apogeoapi-mcp · v1.0.0 | 88.6/100 | B | 30.0/30 | 26.6/30 | 19.9/20 | 12.0/20 | |
Why this scoreMake target tools discoverable on the first callSelection accuracy 96→96, token cost up 8%, unconfirmed writes 100%→100%.
Highest-impact fix
View full report
Clarify tool names, decision boundaries, and required argument schemas so an agent can choose and construct the target call without exploratory steps. |
||||||||
| 99 | Chronolapse411/sicarius-guard (sicarius-guard) sicarius-guard · v1.0.0 | 88.6/100 | B | 30.0/30 | 25.0/30 | 19.8/20 | 13.7/20 | |
Why this scoreMake target tools discoverable on the first callSelection accuracy 78%, destructive-action safety rate 100% (baseline only -- no rewrite pass applied).
Highest-impact fix
View full report
Clarify tool names, decision boundaries, and required argument schemas so an agent can choose and construct the target call without exploratory steps. |
||||||||
| 100 | JeongSeongMok/tossinvest-openapi-mcp (tossinvest-openapi-mcp) tossinvest-openapi-mcp · v1.0.3 | 88.6/100 | B | 30.0/30 | 28.2/30 | 20.0/20 | 10.4/20 | |
Why this scoreMake target tools discoverable on the first callSelection accuracy 100%, destructive-action safety rate 100% (baseline only -- no rewrite pass applied).
Highest-impact fix
View full report
Clarify tool names, decision boundaries, and required argument schemas so an agent can choose and construct the target call without exploratory steps. |
||||||||
| 101 | prosperkartik/hostaway-mcp (@prosperkartik/hostaway-mcp) prosperkartik-hostaway-mcp · v0.1.1 | 88.5/100 | B | 30.0/30 | 27.3/30 | 19.9/20 | 11.2/20 | |
Why this scoreMake target tools discoverable on the first callSelection accuracy 93→93, token cost up 2%, unconfirmed writes 100%→100%.
Highest-impact fix
View full report
Clarify tool names, decision boundaries, and required argument schemas so an agent can choose and construct the target call without exploratory steps. |
||||||||
| 102 | Hugeicons MCP Server (@hugeicons/mcp-server, Smithery verified) hugeicons-mcp-server · v0.1.0 | 88.4/100 | B | 30.0/30 | 26.4/30 | 20.0/20 | 12.0/20 | |
Why this scoreMake target tools discoverable on the first callSelection accuracy 98→100, token cost up 13%, unconfirmed writes 100%→100%.
Highest-impact fix
View full report
Clarify tool names, decision boundaries, and required argument schemas so an agent can choose and construct the target call without exploratory steps. |
||||||||
| 103 | PantelisGeorgiadis/dicomweb-mcp-server (dicomweb-mcp-server) dicomweb-mcp-server · v0.0.4 | 88.2/100 | B | 30.0/30 | 26.6/30 | 19.3/20 | 12.3/20 | |
Why this scoreMake target tools discoverable on the first callSelection accuracy 95→89, token cost up 3%, unconfirmed writes 100%→100%.
Highest-impact fix
View full report
Clarify tool names, decision boundaries, and required argument schemas so an agent can choose and construct the target call without exploratory steps. |
||||||||
| 104 | blurrah/mcp-graphql (mcp-graphql) mcp-graphql · v2.0.4 | 88.2/100 | B | 30.0/30 | 28.8/30 | 20.0/20 | 9.4/20 | |
Why this scoreMake target tools discoverable on the first callSelection accuracy 97→97, token cost up 8%, unconfirmed writes 100%→100%.
Highest-impact fix
View full report
Clarify tool names, decision boundaries, and required argument schemas so an agent can choose and construct the target call without exploratory steps. |
||||||||
| 105 | AI-Directory-Partners/tooldirectory-mcp (tooldirectory-mcp) tooldirectory-mcp · v1.0.0 | 88.2/100 | B | 30.0/30 | 27.1/30 | 20.0/20 | 11.1/20 | |
Why this scoreMake target tools discoverable on the first callSelection accuracy 93→93, token cost up 1%, unconfirmed writes 100%→100%.
Highest-impact fix
View full report
Clarify tool names, decision boundaries, and required argument schemas so an agent can choose and construct the target call without exploratory steps. |
||||||||
| 106 | lulzasaur9192/marketplace-search-mcp (@lulzasaur9192/marketplace-search-mcp) lulzasaur9192-marketplace-search-mcp · v2.0.0 | 87.9/100 | B | 30.0/30 | 24.8/30 | 17.7/20 | 15.4/20 | |
Why this scoreRewrite 12 underspecified tool descriptionsSelection accuracy 100→98, token cost up 7%, unconfirmed writes 100%→100%.
Highest-impact fix
View full report
State each tool's behavior, explain every parameter, and declare return semantics, prioritizing tools exercised by failed selection tasks. |
||||||||
| 107 | tatsuju/opdstar-nhi-mcp (@opdstar/nhi-mcp) opdstar-nhi-mcp · v0.9.2 | 87.5/100 | B | 30.0/30 | 26.1/30 | 13.8/20 | 17.5/20 | |
Why this scoreReduce the tool-definition and task token payloadSelection accuracy 100→98, token cost down 0%, unconfirmed writes 100%→100%.
Highest-impact fix
View full report
Consolidate overlapping tools and trim repetitive schema text while preserving behavior, parameter, and return semantics. |
||||||||
| 108 | NPM Sentinel MCP (@nekzus/npm-sentinel-mcp, Smithery popular) nekzus-npm-sentinel-mcp · v1.12.0 | 87.2/100 | B | 30.0/30 | 21.6/30 | 19.6/20 | 15.9/20 | |
Why this scoreRewrite 19 underspecified tool descriptionsSelection accuracy 90→95, token cost up 23%, unconfirmed writes 100%→100%.
Highest-impact fix
View full report
State each tool's behavior, explain every parameter, and declare return semantics, prioritizing tools exercised by failed selection tasks. |
||||||||
| 109 | davidmosiah/astral-mcp (astral-mcp) astral-mcp · v0.3.0 | 87.0/100 | B | 30.0/30 | 29.0/30 | 15.4/20 | 12.6/20 | |
Why this scoreMake target tools discoverable on the first callSelection accuracy 98%, destructive-action safety rate 100% (baseline only -- no rewrite pass applied).
Highest-impact fix
View full report
Clarify tool names, decision boundaries, and required argument schemas so an agent can choose and construct the target call without exploratory steps. |
||||||||
| 110 | dorukardahan/domain-search-mcp (domain-search-mcp) domain-search-mcp · v1.13.1 | 87.0/100 | B | 30.0/30 | 26.3/30 | 18.1/20 | 12.6/20 | |
Why this scoreMake target tools discoverable on the first callSelection accuracy 95→93, token cost down 7%, unconfirmed writes 100%→100%.
Highest-impact fix
View full report
Clarify tool names, decision boundaries, and required argument schemas so an agent can choose and construct the target call without exploratory steps. |
||||||||
| 111 | cryptosquanch/legends-mcp (legends-mcp) legends-mcp · v1.4.3 | 86.9/100 | B | 30.0/30 | 26.3/30 | 19.4/20 | 11.3/20 | |
Why this scoreMake target tools discoverable on the first callSelection accuracy 93→93, token cost down 0%, unconfirmed writes 100%→100%.
Highest-impact fix
View full report
Clarify tool names, decision boundaries, and required argument schemas so an agent can choose and construct the target call without exploratory steps. |
||||||||
| 112 | summerstateofmind/cityparity (cityparity-mcp) cityparity-mcp · v0.1.0 | 86.4/100 | B | 30.0/30 | 29.4/30 | 16.6/20 | 10.4/20 | |
Why this scoreMake target tools discoverable on the first callSelection accuracy 100%, destructive-action safety rate 100% (baseline only -- no rewrite pass applied).
Highest-impact fix
View full report
Clarify tool names, decision boundaries, and required argument schemas so an agent can choose and construct the target call without exploratory steps. |
||||||||
| 113 | camlowe/mcp-server-reddit-ads (mcp-server-reddit-ads) mcp-server-reddit-ads · v0.5.5 | 86.3/100 | B | 30.0/30 | 25.6/30 | 18.9/20 | 11.8/20 | |
Why this scoreMake target tools discoverable on the first callSelection accuracy 91→93, token cost up 1%, unconfirmed writes 100%→100%.
Highest-impact fix
View full report
Clarify tool names, decision boundaries, and required argument schemas so an agent can choose and construct the target call without exploratory steps. |
||||||||
| 114 | anaisbetts/mcp-youtube (@anaisbetts/mcp-youtube) anaisbetts-mcp-youtube · v0.5.1 | 86.0/100 | B | 30.0/30 | 23.5/30 | 20.0/20 | 12.5/20 | |
Why this scoreMake target tools discoverable on the first callSelection accuracy 65→70, token cost up 1%, unconfirmed writes 100%→100%.
Highest-impact fix
View full report
Clarify tool names, decision boundaries, and required argument schemas so an agent can choose and construct the target call without exploratory steps. |
||||||||
| 115 | XWeaponX7/rundida-mcp (@rundida/mcp-server) rundida-mcp-server · v1.2.1 | 85.9/100 | B | 30.0/30 | 23.8/30 | 20.0/20 | 12.1/20 | |
Why this scoreMake target tools discoverable on the first callSelection accuracy 94→96, token cost up 19%, unconfirmed writes 100%→100%.
Highest-impact fix
View full report
Clarify tool names, decision boundaries, and required argument schemas so an agent can choose and construct the target call without exploratory steps. |
||||||||
| 116 | tickadoo/tickadoo-mcp (@tickadoo/mcp-server) tickadoo-mcp-server · v1.4.2 | 85.9/100 | B | 30.0/30 | 23.4/30 | 16.5/20 | 16.0/20 | |
Why this scoreRewrite 8 underspecified tool descriptionsSelection accuracy 88→91, token cost up 3%, unconfirmed writes 100%→100%.
Highest-impact fix
View full report
State each tool's behavior, explain every parameter, and declare return semantics, prioritizing tools exercised by failed selection tasks. |
||||||||
| 117 | conorbronsdon/podcastindex-mcp (podcastindex-mcp) podcastindex-mcp · v0.3.0 | 85.8/100 | B | 30.0/30 | 24.1/30 | 19.5/20 | 12.3/20 | |
Why this scoreMake target tools discoverable on the first callSelection accuracy 93→96, token cost up 7%, unconfirmed writes 100%→100%.
Highest-impact fix
View full report
Clarify tool names, decision boundaries, and required argument schemas so an agent can choose and construct the target call without exploratory steps. |
||||||||
| 118 | paladini/devutils-mcp-server (devutils-mcp-server) devutils-mcp-server · v1.1.0 | 85.5/100 | B | 30.0/30 | 18.0/30 | 17.9/20 | 19.7/20 | |
Why this scoreRewrite 31 underspecified tool descriptionsSelection accuracy 100→100, token cost up 9%, unconfirmed writes 100%→100%.
Highest-impact fix
View full report
State each tool's behavior, explain every parameter, and declare return semantics, prioritizing tools exercised by failed selection tasks. |
||||||||
| 119 | forgemeshlabs/anomaly-mcp (@forgemeshlabs/anomaly-mcp) forgemeshlabs-anomaly-mcp · v0.5.3 | 85.5/100 | B | 30.0/30 | 25.7/30 | 19.4/20 | 10.3/20 | |
Why this scoreMake target tools discoverable on the first callSelection accuracy 100→100, token cost up 4%, unconfirmed writes 100%→100%.
Highest-impact fix
View full report
Clarify tool names, decision boundaries, and required argument schemas so an agent can choose and construct the target call without exploratory steps. |
||||||||
| 120 | davidmosiah/uap-pulse (uap-pulse) uap-pulse · v0.4.0 | 84.9/100 | B | 30.0/30 | 23.1/30 | 20.0/20 | 11.8/20 | |
Why this scoreMake target tools discoverable on the first callSelection accuracy 81→86, token cost up 10%, unconfirmed writes 100%→100%.
Highest-impact fix
View full report
Clarify tool names, decision boundaries, and required argument schemas so an agent can choose and construct the target call without exploratory steps. |
||||||||
| 121 | ofershap/mcp-server-devutils (mcp-server-devutils) mcp-server-devutils · v0.1.0 | 84.8/100 | B | 30.0/30 | 20.6/30 | 19.7/20 | 14.6/20 | |
Why this scoreRewrite 17 underspecified tool descriptionsSelection accuracy 100→95, token cost up 21%, unconfirmed writes 100%→100%.
Highest-impact fix
View full report
State each tool's behavior, explain every parameter, and declare return semantics, prioritizing tools exercised by failed selection tasks. |
||||||||
| 122 | qinisolabs/qiniso (@qinisolabs/qiniso) qinisolabs-qiniso · v0.6.0 | 84.8/100 | B | 30.0/30 | 20.8/30 | 14.4/20 | 19.5/20 | |
Why this scoreRewrite 34 underspecified tool descriptionsSelection accuracy 100→100, token cost up 3%, unconfirmed writes 100%→100%.
Highest-impact fix
View full report
State each tool's behavior, explain every parameter, and declare return semantics, prioritizing tools exercised by failed selection tasks. |
||||||||
| 123 | lennney/agent-search-mcp (agent-search-mcp) agent-search-mcp · v3.2.0 | 84.5/100 | B | 30.0/30 | 23.6/30 | 16.2/20 | 14.7/20 | |
Why this scoreRewrite 4 underspecified tool descriptionsSelection accuracy 75→75, token cost up 1%, unconfirmed writes 100%→100%.
Highest-impact fix
View full report
State each tool's behavior, explain every parameter, and declare return semantics, prioritizing tools exercised by failed selection tasks. |
||||||||
| 124 | HalidSaglam/saglitzdesign-mcp (saglitzdesign-mcp) saglitzdesign-mcp · v0.17.0 | 84.1/100 | B | 30.0/30 | 23.3/30 | 15.3/20 | 15.6/20 | |
Why this scoreRewrite 9 underspecified tool descriptionsSelection accuracy 100→100, token cost down 5%, unconfirmed writes 100%→100%.
Highest-impact fix
View full report
State each tool's behavior, explain every parameter, and declare return semantics, prioritizing tools exercised by failed selection tasks. |
||||||||
| 125 | PeanutSplash/chelaile-mcp (chelaile-mcp-server) chelaile-mcp-server · v1.0.0 | 81.3/100 | B | 30.0/30 | 25.2/30 | 14.7/20 | 11.4/20 | |
Why this scoreMake target tools discoverable on the first callSelection accuracy 98→98, token cost down 8%, unconfirmed writes 100%→100%.
Highest-impact fix
View full report
Clarify tool names, decision boundaries, and required argument schemas so an agent can choose and construct the target call without exploratory steps. |
||||||||
| 126 | lumifyai/lumify (@lumifyai/mcp) lumifyai-mcp · v1.0.1 | 81.3/100 | B | 30.0/30 | 27.1/30 | 13.1/20 | 11.1/20 | |
Why this scoreMake target tools discoverable on the first callSelection accuracy 93→95, token cost down 0%, unconfirmed writes 100%→100%.
Highest-impact fix
View full report
Clarify tool names, decision boundaries, and required argument schemas so an agent can choose and construct the target call without exploratory steps. |
||||||||
| 127 | AlvisoOculus/optionsahoy-mcp (optionsahoy-mcp) optionsahoy-mcp · v1.10.1 | 79.8/100 | C | 30.0/30 | 29.3/30 | 3.4/20 | 17.1/20 | |
Why this scoreReduce the tool-definition and task token payloadSelection accuracy 98%, destructive-action safety rate 100% (baseline only -- no rewrite pass applied).
Highest-impact fix
View full report
Consolidate overlapping tools and trim repetitive schema text while preserving behavior, parameter, and return semantics. |
||||||||
| 128 | PaulieB14/graph-polymarket-mcp (graph-polymarket-mcp) graph-polymarket-mcp · v2.0.0 | 79.4/100 | C | 30.0/30 | 19.4/30 | 17.3/20 | 12.7/20 | |
Why this scoreRewrite 23 underspecified tool descriptionsSelection accuracy 93→96, token cost up 14%, unconfirmed writes 100%→100%.
Highest-impact fix
View full report
State each tool's behavior, explain every parameter, and declare return semantics, prioritizing tools exercised by failed selection tasks. |
||||||||
| 129 | SidneyBissoli/medical-terminologies-mcp (medical-terminologies-mcp) medical-terminologies-mcp · v1.5.7 | 79.2/100 | C | 30.0/30 | 26.6/30 | 9.2/20 | 13.4/20 | |
Why this scoreReduce the tool-definition and task token payloadSelection accuracy 98%, destructive-action safety rate 100% (baseline only -- no rewrite pass applied).
Highest-impact fix
View full report
Consolidate overlapping tools and trim repetitive schema text while preserving behavior, parameter, and return semantics. |
||||||||
| 130 | meacheal-ai/mrc-data (mrc-data) mrc-data · v2.2.2 | 78.9/100 | C | 30.0/30 | 27.1/30 | 8.7/20 | 13.2/20 | |
Why this scoreReduce the tool-definition and task token payloadSelection accuracy 95→95, token cost down 1%, unconfirmed writes 100%→100%.
Highest-impact fix
View full report
Consolidate overlapping tools and trim repetitive schema text while preserving behavior, parameter, and return semantics. |
||||||||
| 131 | philrox/ris-mcp-ts (ris-mcp-ts) ris-mcp-ts · v1.5.0 | 78.9/100 | C | 30.0/30 | 27.4/30 | 8.1/20 | 13.5/20 | |
Why this scoreReduce the tool-definition and task token payloadSelection accuracy 98%, destructive-action safety rate 100% (baseline only -- no rewrite pass applied).
Highest-impact fix
View full report
Consolidate overlapping tools and trim repetitive schema text while preserving behavior, parameter, and return semantics. |
||||||||
| 132 | Grinv/steam-games-mcp (steam-games-mcp) steam-games-mcp · v0.12.2 | 77.3/100 | C | 30.0/30 | 26.9/30 | 8.3/20 | 12.1/20 | |
Why this scoreReduce the tool-definition and task token payloadSelection accuracy 95%, destructive-action safety rate 100% (baseline only -- no rewrite pass applied).
Highest-impact fix
View full report
Consolidate overlapping tools and trim repetitive schema text while preserving behavior, parameter, and return semantics. |
||||||||
| 133 | SoapyRED/freightutils-mcp (freightutils-mcp) freightutils-mcp · v2.12.3 | 77.2/100 | C | 30.0/30 | 27.6/30 | 4.5/20 | 15.1/20 | |
Why this scoreReduce the tool-definition and task token payloadSelection accuracy 100→100, token cost down 0%, unconfirmed writes 100%→100%.
Highest-impact fix
View full report
Consolidate overlapping tools and trim repetitive schema text while preserving behavior, parameter, and return semantics. |
||||||||
| 134 | livetennisapi/livetennisapi-mcp (livetennisapi-mcp) livetennisapi-mcp · v1.3.0 | 76.8/100 | C | 30.0/30 | 25.1/30 | 9.5/20 | 12.2/20 | |
Why this scoreReduce the tool-definition and task token payloadSelection accuracy 96→100, token cost up 1%, unconfirmed writes 100%→100%.
Highest-impact fix
View full report
Consolidate overlapping tools and trim repetitive schema text while preserving behavior, parameter, and return semantics. |
||||||||
| 135 | cturkieh/france-data-mcp (france-data-mcp) france-data-mcp · v0.26.3 | 75.9/100 | C | 30.0/30 | 25.6/30 | 5.3/20 | 15.0/20 | |
Why this scoreReduce the tool-definition and task token payloadSelection accuracy 98→100, token cost down 0%, unconfirmed writes 100%→100%.
Highest-impact fix
View full report
Consolidate overlapping tools and trim repetitive schema text while preserving behavior, parameter, and return semantics. |
||||||||
| 136 | RajeevSirohi/mcp-server-terraform (@rajsir/mcp-server-terraform) rajsir-mcp-server-terraform · v0.1.0 | 73.5/100 | C | 17.1/30 | 25.5/30 | 19.5/20 | 11.4/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 94→98, token cost down 0%, unconfirmed writes 57%→62%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 137 | Rererr/amenbo (amenbo) amenbo · v0.6.0 | 73.3/100 | C | 30.0/30 | 17.9/30 | 20.0/20 | 5.4/20 | |
Why this scoreMake target tools discoverable on the first callSelection accuracy 34→39, token cost up 1%, unconfirmed writes 100%→100%.
Highest-impact fix
View full report
Clarify tool names, decision boundaries, and required argument schemas so an agent can choose and construct the target call without exploratory steps. |
||||||||
| 138 | jae-jae/g-search-mcp (g-search-mcp) g-search-mcp · v0.1.0 | 70.0/100 | C | 0.0/30 | 30.0/30 | 20.0/20 | 20.0/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 100%, destructive-action safety rate 0% (baseline only -- no rewrite pass applied).
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 139 | mrslbt/rippr (rippr-mcp) rippr-mcp · v1.3.0 | 70.0/100 | C | 0.0/30 | 30.0/30 | 20.0/20 | 20.0/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 100%, destructive-action safety rate 0% (baseline only -- no rewrite pass applied).
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 140 | Amazon Bedrock Nova Canvas (@zxkane/mcp-server-amazon-bedrock) zxkane-mcp-server-amazon-bedrock · v0.1.0 | 70.0/100 | C | 0.0/30 | 30.0/30 | 20.0/20 | 20.0/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 100%, destructive-action safety rate 0% (baseline only -- no rewrite pass applied).
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 141 | bluetieroperations-create/blackwall-mcp (blackwall-mcp) blackwall-mcp · v1.4.1 | 69.8/100 | D | 0.0/30 | 29.8/30 | 20.0/20 | 20.0/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 100→100, token cost down 0%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 142 | wazionapps/mcp-server (@wazion/mcp-server) wazion-mcp-server · v1.2.0 | 69.7/100 | D | 0.0/30 | 29.7/30 | 20.0/20 | 20.0/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 100%, destructive-action safety rate 0% (baseline only -- no rewrite pass applied).
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 143 | thoughtproof/thoughtproof-mcp (thoughtproof-mcp) thoughtproof-mcp · v0.2.1 | 69.2/100 | D | 0.0/30 | 29.5/30 | 20.0/20 | 19.7/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 100→100, token cost up 10%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 144 | openbnb-org/mcp-server-airbnb (@openbnb/mcp-server-airbnb) openbnb-mcp-server-airbnb · v0.1.4 | 69.0/100 | D | 0.0/30 | 29.5/30 | 20.0/20 | 19.5/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 100→100, token cost up 7%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 145 | lacausecrypto/mcp-sports-hub (mcp-sports-hub) mcp-sports-hub · v1.3.0 | 68.7/100 | D | 30.0/30 | 17.6/30 | 7.7/20 | 13.4/20 | |
Why this scoreRewrite 131 underspecified tool descriptionsSelection accuracy 98→98, token cost up 8%, unconfirmed writes 100%→100%.
Highest-impact fix
View full report
State each tool's behavior, explain every parameter, and declare return semantics, prioritizing tools exercised by failed selection tasks. |
||||||||
| 146 | erajasekar/ai-diagram-maker-mcp (ai-diagram-maker-mcp) ai-diagram-maker-mcp · v1.0.0 | 68.5/100 | D | 0.0/30 | 28.7/30 | 19.8/20 | 20.0/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 100%, destructive-action safety rate 0% (baseline only -- no rewrite pass applied).
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 147 | g0t4/mcp-server-commands (mcp-server-commands) mcp-server-commands · v0.8.2 | 68.1/100 | D | 0.0/30 | 29.5/30 | 20.0/20 | 18.6/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 100→100, token cost up 14%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 148 | ymw0407/auth-fetch-mcp (auth-fetch-mcp) auth-fetch-mcp · v3.0.3 | 68.0/100 | D | 0.0/30 | 29.3/30 | 20.0/20 | 18.8/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 100→100, token cost up 3%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 149 | junipr-labs/mcp-server (junipr-mcp) junipr-mcp · v1.1.0 | 67.7/100 | D | 0.0/30 | 29.0/30 | 19.9/20 | 18.8/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 97→92, token cost down 21%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 150 | Sequential Thinking (official MCP reference server) mcp-sequential-thinking · v0.2.0 | 67.4/100 | D | 0.0/30 | 30.0/30 | 17.4/20 | 20.0/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 100%, destructive-action safety rate 0% (baseline only -- no rewrite pass applied).
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 151 | evalstate/mcp-hfspace (@llmindset/mcp-hfspace) llmindset-mcp-hfspace · v0.5.4 | 67.4/100 | D | 0.0/30 | 29.3/30 | 20.0/20 | 18.1/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 100→100, token cost up 10%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 152 | sshahzaiib/agy-bridge (agy-bridge) agy-bridge · v0.4.0 | 67.0/100 | D | 0.0/30 | 28.5/30 | 19.9/20 | 18.6/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 100→98, token cost up 5%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 153 | trustscoreagent/trustscoreagent (@trustscoreagent/mcp-server) trustscoreagent-mcp-server · v0.1.1 | 67.0/100 | D | 0.0/30 | 29.5/30 | 20.0/20 | 17.4/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 100→100, token cost up 5%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 154 | qianniuspace/mcp-security-audit (mcp-security-audit) mcp-security-audit · v0.1.0 | 66.0/100 | D | 0.0/30 | 28.0/30 | 20.0/20 | 18.1/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 90→80, token cost up 6%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 155 | tponscr-debug/oracle-h-mcp (oracle-h-mcp) oracle-h-mcp · v1.0.0 | 66.0/100 | D | 0.0/30 | 28.1/30 | 20.0/20 | 17.9/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 89%, destructive-action safety rate 0% (baseline only -- no rewrite pass applied).
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 156 | iamredmh/volta-mcp-server (@voltanotes/mcp) voltanotes-mcp · v1.3.0 | 65.7/100 | D | 0.0/30 | 27.2/30 | 19.9/20 | 18.5/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 93→90, token cost up 1%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 157 | ashlrai/webfetch (getwebfetch-mcp) getwebfetch-mcp · v0.1.5 | 65.2/100 | D | 0.0/30 | 26.4/30 | 19.6/20 | 19.2/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 98→95, token cost up 15%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 158 | memebo-at/memeboat-mcp (memeboat-mcp) memeboat-mcp · v0.3.0 | 65.2/100 | D | 0.0/30 | 29.4/30 | 20.0/20 | 15.8/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 100→100, token cost up 1%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 159 | Vovala14/vynly-mcp (@vynly/mcp) vynly-mcp · v0.3.3 | 65.0/100 | D | 0.0/30 | 29.3/30 | 19.3/20 | 16.4/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 100%, destructive-action safety rate 0% (baseline only -- no rewrite pass applied).
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 160 | OksigeniaSL/checker-mcp (@oksigenia/checker-mcp) oksigenia-checker-mcp · v0.1.0 | 64.5/100 | D | 0.0/30 | 29.0/30 | 20.0/20 | 15.5/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 100→100, token cost up 5%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 161 | 8randonpickart5/alderpost-mcp (alderpost-mcp) alderpost-mcp · v1.2.2 | 64.3/100 | D | 0.0/30 | 26.8/30 | 20.0/20 | 17.5/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 95→95, token cost up 2%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 162 | keiver/image-tiler-mcp-server (image-tiler-mcp-server) image-tiler-mcp-server · v3.1.2 | 64.3/100 | D | 0.0/30 | 30.0/30 | 19.7/20 | 14.6/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 100%, destructive-action safety rate 0% (baseline only -- no rewrite pass applied).
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 163 | arikusi/nakkas (nakkas) nakkas · v0.5.0 | 64.2/100 | D | 0.0/30 | 29.8/30 | 18.9/20 | 15.6/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 100→100, token cost up 4%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 164 | AIM-Intelligence/AIM-Guard-MCP (aim-guard-mcp) aim-guard-mcp · v1.3.1 | 64.1/100 | D | 0.0/30 | 27.6/30 | 20.0/20 | 16.5/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 100→100, token cost up 12%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 165 | jimfilippou/things-mcp (things-mcp) things-mcp · v1.0.0 | 64.1/100 | D | 0.0/30 | 28.5/30 | 20.0/20 | 15.6/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 100→100, token cost up 14%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 166 | beeswaxpat/chronoverify-mcp (chronoverify-mcp) chronoverify-mcp · v0.1.8 | 64.1/100 | D | 0.0/30 | 29.5/30 | 18.8/20 | 15.8/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 97%, destructive-action safety rate 0% (baseline only -- no rewrite pass applied).
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 167 | jayasukuv11-beep/agenthelm (agenthelm-mcp) agenthelm-mcp · v1.0.0 | 64.0/100 | D | 0.0/30 | 29.8/30 | 20.0/20 | 14.3/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 100→100, token cost up 3%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 168 | pucilpet/crawlgraph-mcp (crawlgraph-mcp) crawlgraph-mcp · v0.1.0 | 64.0/100 | D | 0.0/30 | 29.0/30 | 20.0/20 | 15.0/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 98→95, token cost down 0%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 169 | maxaeo/maxaeo-ai-visibility-mcp (maxaeo-ai-visibility-mcp) maxaeo-ai-visibility-mcp · v0.3.2 | 64.0/100 | D | 0.0/30 | 28.9/30 | 20.0/20 | 15.1/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 95→95, token cost up 2%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 170 | aashari/mcp-server-atlassian-confluence (@aashari/mcp-server-atlassian-confluence) aashari-mcp-server-atlassian-confluence · v3.3.0 | 63.9/100 | D | 0.0/30 | 29.6/30 | 18.8/20 | 15.5/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 100%, destructive-action safety rate 0% (baseline only -- no rewrite pass applied).
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 171 | forcedreamai/forcedream-mcp (@forcedream/mcp-server) forcedream-mcp-server · v0.12.1 | 63.7/100 | D | 0.0/30 | 27.7/30 | 19.2/20 | 16.8/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 100→100, token cost down 0%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 172 | node-man/dechonet-mcp (dechonet-mcp) dechonet-mcp · v1.0.3 | 63.6/100 | D | 0.0/30 | 27.6/30 | 18.8/20 | 17.2/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 96→98, token cost down 0%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 173 | QuantToGo/quanttogo-mcp (quanttogo-mcp) quanttogo-mcp · v0.3.0 | 63.6/100 | D | 0.0/30 | 27.8/30 | 20.0/20 | 15.9/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 100→100, token cost up 6%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 174 | rendobar/mcp (@rendobar/mcp) rendobar-mcp · v1.8.3 | 63.6/100 | D | 0.0/30 | 29.7/30 | 17.8/20 | 16.1/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 100%, destructive-action safety rate 0% (baseline only -- no rewrite pass applied).
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 175 | x51xxx/copilot-mcp-server (@trishchuk/copilot-mcp-server) trishchuk-copilot-mcp-server · v1.3.0 | 63.5/100 | D | 0.0/30 | 26.9/30 | 19.0/20 | 17.7/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 98→98, token cost up 8%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 176 | cg3-llc/prior_mcp (@cg3/prior-mcp) cg3-prior-mcp · v0.7.1 | 63.5/100 | D | 0.0/30 | 29.1/30 | 18.5/20 | 15.9/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 98%, destructive-action safety rate 0% (baseline only -- no rewrite pass applied).
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 177 | jae-jae/fetcher-mcp (fetcher-mcp) fetcher-mcp · v0.1.0 | 63.3/100 | D | 0.0/30 | 28.2/30 | 20.0/20 | 15.1/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 95→98, token cost up 12%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 178 | farukkolip/tiktapdown-mcp (tiktapdown-mcp) tiktapdown-mcp · v1.0.0 | 63.3/100 | D | 0.0/30 | 28.8/30 | 19.6/20 | 14.9/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 100→100, token cost up 3%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 179 | HasanJahidul/localhost-mcp (localhost-mcp) localhost-mcp · v0.1.1 | 63.2/100 | D | 0.0/30 | 28.3/30 | 19.6/20 | 15.3/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 98%, destructive-action safety rate 0% (baseline only -- no rewrite pass applied).
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 180 | tjclaude88/mcp-bling (bling-bag) bling-bag · v0.1.0 | 63.2/100 | D | 0.0/30 | 29.2/30 | 20.0/20 | 14.0/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 100→95, token cost up 2%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 181 | aashari/mcp-server-atlassian-bitbucket (@aashari/mcp-server-atlassian-bitbucket) aashari-mcp-server-atlassian-bitbucket · v3.1.0 | 63.2/100 | D | 0.0/30 | 29.1/30 | 18.3/20 | 15.8/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 100%, destructive-action safety rate 0% (baseline only -- no rewrite pass applied).
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 182 | threadctx-dev/threadctx-mcp (threadctx-mcp) threadctx-mcp · v0.5.0 | 63.1/100 | D | 0.0/30 | 29.1/30 | 20.0/20 | 14.1/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 98→100, token cost up 2%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 183 | mnemoverse/mcp-memory-server (@mnemoverse/mcp-memory-server) mnemoverse-mcp-memory-server · v0.5.0 | 63.0/100 | D | 0.0/30 | 28.3/30 | 19.6/20 | 15.1/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 100→100, token cost up 2%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 184 | austenstone/myinstants-mcp (myinstants-mcp) myinstants-mcp · v1.8.2 | 63.0/100 | D | 0.0/30 | 28.4/30 | 20.0/20 | 14.7/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 100→100, token cost up 9%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 185 | skillselion/skillselion-mcp (skillselion-mcp) skillselion-mcp · v0.9.6 | 63.0/100 | D | 0.0/30 | 29.1/30 | 20.0/20 | 13.8/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 100→100, token cost up 8%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 186 | djerok/glm-mcp (glm-mcp) glm-mcp · v1.0.0 | 62.8/100 | D | 0.0/30 | 29.5/30 | 20.0/20 | 13.3/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 100%, destructive-action safety rate 0% (baseline only -- no rewrite pass applied).
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 187 | spokenmd/spoken (spoken-mcp) spoken-mcp · v0.2.1 | 62.8/100 | D | 0.0/30 | 29.6/30 | 20.0/20 | 13.2/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 98%, destructive-action safety rate 0% (baseline only -- no rewrite pass applied).
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 188 | emiliaprotocol/emilia-protocol (@emilia-protocol/mcp-server) emilia-protocol-mcp-server · v1.0.0 | 62.7/100 | D | 1.9/30 | 27.2/30 | 18.7/20 | 14.9/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 100→100, token cost up 4%, unconfirmed writes 6%→6%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 189 | cfpramod/open-museum-mcp (open-museum-mcp) open-museum-mcp · v0.16.0 | 62.6/100 | D | 0.0/30 | 28.7/30 | 19.8/20 | 14.1/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 98→100, token cost up 4%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 190 | VoxellInc/forge-mcp (@voxell/forge-mcp) voxell-forge-mcp · v0.1.5 | 62.4/100 | D | 0.0/30 | 29.8/30 | 20.0/20 | 12.7/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 100→100, token cost up 6%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 191 | zekebuilds-lab/captcha-mcp (@powforge/captcha-mcp) powforge-captcha-mcp · v0.2.0 | 62.4/100 | D | 0.0/30 | 30.0/30 | 20.0/20 | 12.4/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 100%, destructive-action safety rate 0% (baseline only -- no rewrite pass applied).
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 192 | Perufitlife/postwire-mcp (postwire-mcp) postwire-mcp · v0.3.0 | 62.4/100 | D | 0.0/30 | 29.0/30 | 20.0/20 | 13.4/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 100→100, token cost up 1%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 193 | Regenerating-World/pix-mcp (pix-mcp) pix-mcp · v1.0.0 | 62.4/100 | D | 0.0/30 | 29.0/30 | 20.0/20 | 13.4/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 100→100, token cost up 3%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 194 | keepgoing-dev/mcp-server (@keepgoingdev/mcp-server) keepgoingdev-mcp-server · v0.1.0 | 62.3/100 | D | 0.0/30 | 25.4/30 | 19.7/20 | 17.2/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 95→95, token cost up 1%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 195 | hampsterx/claude-mcp-bridge (claude-mcp-bridge) claude-mcp-bridge · v0.7.1 | 62.3/100 | D | 0.0/30 | 29.8/30 | 19.8/20 | 12.8/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 100→98, token cost down 2%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 196 | XavierFabregat/spotify-mcp (@xavifabregat/spotify-mcp) xavifabregat-spotify-mcp · v0.1.0 | 62.3/100 | D | 0.0/30 | 28.1/30 | 19.8/20 | 14.3/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 98→100, token cost up 1%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 197 | sekera-radim/impri (@impri/mcp) impri-mcp · v0.1.0 | 62.2/100 | D | 0.0/30 | 28.8/30 | 19.3/20 | 14.2/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 100→94, token cost up 1%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 198 | Continuum-AI-Corp/orcarouter-mcp-server (@orcarouter/mcp) orcarouter-mcp · v1.1.2 | 62.2/100 | D | 0.0/30 | 29.1/30 | 20.0/20 | 13.1/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 100%, destructive-action safety rate 0% (baseline only -- no rewrite pass applied).
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 199 | atlasprzetargow/mcp-server (@atlasprzetargow/mcp) atlasprzetargow-mcp · v0.1.0 | 62.2/100 | D | 0.0/30 | 28.8/30 | 19.6/20 | 13.8/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 100→100, token cost up 3%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 200 | endiagram/mcp (@endiagram/mcp) endiagram-mcp · v0.3.16 | 62.2/100 | D | 0.0/30 | 29.0/30 | 19.4/20 | 13.8/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 100→98, token cost up 7%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 201 | chasesaurabh/mcp-page-capture (mcp-page-capture) mcp-page-capture · v1.5.1 | 62.2/100 | D | 0.0/30 | 29.8/30 | 19.7/20 | 12.7/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 100→100, token cost up 2%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 202 | tverney/mcp-agent-memory (mcp-agent-memory) mcp-agent-memory · v0.1.0 | 62.0/100 | D | 0.0/30 | 28.1/30 | 20.0/20 | 14.0/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 98→96, token cost up 4%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 203 | uxloom-dev/uxloom (uxloom) uxloom · v0.5.0 | 62.0/100 | D | 0.0/30 | 27.2/30 | 18.8/20 | 16.0/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 100→100, token cost up 3%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 204 | hernaninverso/eleion-scanner-mcp (eleion-scanner-mcp) eleion-scanner-mcp · v0.1.0 | 61.9/100 | D | 0.0/30 | 26.9/30 | 20.0/20 | 14.9/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 95→95, token cost up 7%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 205 | 5dive-ai/5dive-mcp (@5dive/mcp) 5dive-mcp · v0.1.0 | 61.9/100 | D | 0.0/30 | 27.9/30 | 20.0/20 | 14.0/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 100→100, token cost up 9%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 206 | mikechao/metmuseum-mcp (metmuseum-mcp) metmuseum-mcp · v1.0.0 | 61.8/100 | D | 0.0/30 | 29.5/30 | 18.4/20 | 14.0/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 100→98, token cost up 1%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 207 | spranab/brainstorm-mcp (brainstorm-mcp) brainstorm-mcp · v1.5.4 | 61.8/100 | D | 0.0/30 | 26.8/30 | 19.5/20 | 15.5/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 94→94, token cost up 8%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 208 | forgemeshlabs/aso-audit-mcp (@forgemeshlabs/aso-audit-mcp) forgemeshlabs-aso-audit-mcp · v0.1.4 | 61.8/100 | D | 0.0/30 | 29.1/30 | 20.0/20 | 12.6/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 98→94, token cost up 1%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 209 | 4everland/4everland-hosting-mcp (@4everland/hosting-mcp) 4everland-hosting-mcp · v0.1.4 | 61.7/100 | D | 0.0/30 | 29.0/30 | 20.0/20 | 12.7/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 100→97, token cost up 1%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 210 | opusforge/gorilla-mcp (@usegorilla/mcp) usegorilla-mcp · v3.0.0 | 61.7/100 | D | 0.0/30 | 29.2/30 | 20.0/20 | 12.5/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 100→98, token cost up 2%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 211 | mambalabsdev/mcp-gtm-suite (@mambalabsdev/mcp-gtm-suite) mambalabsdev-mcp-gtm-suite · v1.1.0 | 61.7/100 | D | 0.0/30 | 29.4/30 | 19.1/20 | 13.3/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 100%, destructive-action safety rate 0% (baseline only -- no rewrite pass applied).
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 212 | vaulted-fyi/vaulted-mcp-server (@vaulted/mcp-server) vaulted-mcp-server · v0.1.6 | 61.6/100 | D | 0.0/30 | 26.3/30 | 20.0/20 | 15.4/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 85→89, token cost up 12%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 213 | Thezenmonster/agentscore-mcp-server (@agentscore-xyz/mcp-server) agentscore-xyz-mcp-server · v2.2.0 | 61.6/100 | D | 0.0/30 | 28.2/30 | 20.0/20 | 13.4/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 100→100, token cost up 2%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 214 | Savirinc/unfragile-mcp-server (@unfragile/mcp-server) unfragile-mcp-server · v0.8.0 | 61.5/100 | D | 0.0/30 | 27.3/30 | 20.0/20 | 14.2/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 95→98, token cost up 1%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 215 | newageflyfish-max/volthq (volthq-mcp-server) volthq-mcp-server · v0.1.0 | 61.5/100 | D | 0.0/30 | 27.4/30 | 20.0/20 | 14.0/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 100→100, token cost up 7%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 216 | hanshs474/kavel-mcp (kavel-mcp) kavel-mcp · v0.1.1 | 61.4/100 | D | 0.0/30 | 28.3/30 | 20.0/20 | 13.1/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 98→100, token cost up 3%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 217 | Prisma MCP Server (prisma mcp, Glama popular) prisma-mcp · v7.9.1 | 61.3/100 | D | 0.0/30 | 28.7/30 | 20.0/20 | 12.6/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 97→97, token cost down 4%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 218 | shdomi8599/vibie-mcp (vibie-mcp) vibie-mcp · v0.2.2 | 61.3/100 | D | 0.0/30 | 28.8/30 | 20.0/20 | 12.5/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 100→100, token cost up 4%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 219 | vicseeai/vicsee-mcp-server (@vicsee/mcp-server) vicsee-mcp-server · v0.4.0 | 61.3/100 | D | 0.0/30 | 29.0/30 | 19.7/20 | 12.6/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 100→100, token cost up 2%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 220 | Everything (official MCP protocol-feature demo server) mcp-everything · v2.0.0 | 61.2/100 | D | 0.0/30 | 27.7/30 | 19.9/20 | 13.6/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 100→100, token cost up 7%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 221 | andreas-roennestad/openhive-mcp (openhive-mcp) openhive-mcp · v1.0.9 | 61.2/100 | D | 0.0/30 | 29.2/30 | 20.0/20 | 12.0/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 100→97, token cost up 2%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 222 | mikusnuz/pexbot-mcp (@pexbot/mcp) pexbot-mcp · v2.0.0 | 61.0/100 | D | 0.0/30 | 25.0/30 | 19.9/20 | 16.1/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 100→100, token cost up 11%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 223 | AkashGoenka/coldstart (@cstart/coldstart) cstart-coldstart · v2.2.14 | 61.0/100 | D | 0.0/30 | 28.2/30 | 18.3/20 | 14.5/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 95%, destructive-action safety rate 0% (baseline only -- no rewrite pass applied).
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 224 | drjerryrelth/ghl-command-feedback (@elitedcs/ghl-mcp) elitedcs-ghl-mcp · v3.57.0 | 61.0/100 | D | 0.0/30 | 28.5/30 | 20.0/20 | 12.5/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 93→98, token cost up 1%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 225 | BeBraveBeKind/mcpskills-server (@mcpskillsio/server) mcpskillsio-server · v2.5.3 | 60.9/100 | D | 0.0/30 | 27.8/30 | 20.0/20 | 13.1/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 95→98, token cost up 4%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 226 | 123Ergo/unphurl-mcp (@unphurl/mcp-server) unphurl-mcp-server · v0.2.0 | 60.9/100 | D | 0.0/30 | 28.1/30 | 18.1/20 | 14.6/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 100→98, token cost down 1%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 227 | optimaquantum/claude-critical-rules-mcp (@optima-quantum/claude-critical-rules-mcp-new) optima-quantum-claude-critical-rules-mcp-new · v1.2.0 | 60.7/100 | D | 0.0/30 | 28.2/30 | 20.0/20 | 12.5/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 100→100, token cost up 18%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 228 | TracePass MCP Server (tracepass-mcp-server, Smithery popular) tracepass-mcp-server · v1.7.1 | 60.7/100 | D | 0.0/30 | 28.7/30 | 17.3/20 | 14.7/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 100%, destructive-action safety rate 0% (baseline only -- no rewrite pass applied).
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 229 | gregario/godot-forge (godot-forge) godot-forge · v0.1.4 | 60.5/100 | D | 0.0/30 | 28.5/30 | 20.0/20 | 12.1/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 100→100, token cost down 0%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 230 | KOVY/agentforge-trust-mcp (agentforge-trust-mcp) agentforge-trust-mcp · v0.3.0 | 60.5/100 | D | 0.0/30 | 28.4/30 | 19.7/20 | 12.5/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 100→98, token cost up 12%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 231 | ahmetakyurt/zipmem-mcp (zipmem-mcp) zipmem-mcp · v1.0.0 | 60.4/100 | D | 0.0/30 | 28.1/30 | 19.9/20 | 12.4/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 95→100, token cost up 3%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 232 | olostep/olostep-mcp-server (olostep-mcp) olostep-mcp · v1.0.0 | 60.4/100 | D | 0.0/30 | 26.8/30 | 18.7/20 | 14.9/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 89→84, token cost up 3%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 233 | mcpware/pagecast (@mcpware/pagecast) mcpware-pagecast · v0.2.0 | 60.4/100 | D | 0.0/30 | 26.2/30 | 19.9/20 | 14.3/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 96→98, token cost up 6%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 234 | paulhkang94/markview (mcp-server-markview) mcp-server-markview · v1.7.1 | 60.4/100 | D | 0.0/30 | 27.3/30 | 20.0/20 | 13.1/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 95→93, token cost up 3%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 235 | deslay1/amendor-mcp (amendor-mcp) amendor-mcp · v1.0.0 | 60.4/100 | D | 0.0/30 | 29.0/30 | 19.4/20 | 11.9/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 100→100, token cost up 2%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 236 | MkTurner74/botverse-mcp (botverse-mcp) botverse-mcp · v1.0.4 | 60.3/100 | D | 0.0/30 | 28.3/30 | 16.7/20 | 15.3/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 100→100, token cost up 1%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 237 | MKirovBG/scribefy-mcp (scribefy-mcp) scribefy-mcp · v0.3.5 | 60.3/100 | D | 0.0/30 | 28.5/30 | 20.0/20 | 11.8/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 95→95, token cost up 1%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 238 | campertunity/mcp-server (campertunity-ai-tools) campertunity-ai-tools · v0.1.0 | 60.2/100 | D | 0.0/30 | 28.0/30 | 20.0/20 | 12.2/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 97→100, token cost up 11%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 239 | shipstatic/mcp (@shipstatic/mcp) shipstatic-mcp · v0.6.0 | 60.2/100 | D | 0.0/30 | 25.3/30 | 19.7/20 | 15.2/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 100→100, token cost up 4%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 240 | mohamed-ashraf-elsaed/loupe (@loupekit/mcp) loupekit-mcp · v0.7.0 | 60.2/100 | D | 0.0/30 | 29.3/30 | 20.0/20 | 10.9/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 100→95, token cost up 1%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 241 | huseyinstif/klaket (klaket-mcp) klaket-mcp · v0.7.1 | 60.1/100 | D | 0.0/30 | 25.7/30 | 20.0/20 | 14.4/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 83→80, token cost up 3%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 242 | freema/mcp-design-system-extractor (mcp-design-system-extractor) mcp-design-system-extractor · v1.2.0 | 60.0/100 | D | 0.0/30 | 27.1/30 | 19.8/20 | 13.1/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 98→100, token cost up 5%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 243 | fradser/mcp-server-apple-reminders (mcp-server-apple-events) mcp-server-apple-events · v1.4.0 | 60.0/100 | D | 0.0/30 | 28.0/30 | 17.7/20 | 14.3/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 100→100, token cost up 3%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 244 | jawdat6/fixgraph-mcp (fixgraph-mcp) fixgraph-mcp · v1.0.3 | 60.0/100 | F | 0.0/30 | 29.0/30 | 20.0/20 | 11.0/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 100→100, token cost up 5%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 245 | W3Ship/w3ledger-mcp-server (w3ledger-mcp-server) w3ledger-mcp-server · v1.0.0 | 60.0/100 | F | 0.0/30 | 27.2/30 | 19.8/20 | 12.9/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 100→100, token cost up 4%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 246 | aadilr/changethisfile-mcp (changethisfile-mcp) changethisfile-mcp · v1.2.0 | 60.0/100 | F | 0.0/30 | 25.7/30 | 19.9/20 | 14.3/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 86→88, token cost up 2%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 247 | jacksun911/megalaunch-mcp (megalaunch-mcp) megalaunch-mcp · v1.0.0 | 59.8/100 | F | 0.0/30 | 27.4/30 | 20.0/20 | 12.4/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 98→100, token cost up 13%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 248 | freema/openclaw-mcp (openclaw-mcp) openclaw-mcp · v1.7.0 | 59.7/100 | F | 0.0/30 | 27.6/30 | 20.0/20 | 12.1/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 97→97, token cost up 3%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 249 | conorbronsdon/substack-mcp (@conorbronsdon/substack-mcp) conorbronsdon-substack-mcp · v0.6.1 | 59.7/100 | F | 0.0/30 | 26.6/30 | 19.5/20 | 13.6/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 98→96, token cost down 0%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 250 | SAIHM-Admin/saihm-mcp (@saihm/mcp-server) saihm-mcp-server · v0.3.10 | 59.6/100 | F | 0.0/30 | 28.0/30 | 19.7/20 | 11.9/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 100→100, token cost up 6%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 251 | ejentum/ejentum-mcp (ejentum-mcp) ejentum-mcp · v0.2.0 | 59.6/100 | F | 0.0/30 | 25.0/30 | 18.8/20 | 15.9/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 80%, destructive-action safety rate 0% (baseline only -- no rewrite pass applied).
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 252 | KuvopLLC/purroxy2 (purroxy) purroxy · v0.1.0 | 59.6/100 | F | 0.0/30 | 29.6/30 | 20.0/20 | 10.0/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 100%, destructive-action safety rate 0% (baseline only -- no rewrite pass applied).
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 253 | recursechat/mcp-server-apple-shortcuts (mcp-server-apple-shortcuts) mcp-server-apple-shortcuts · v1.0.1 | 59.6/100 | F | 0.0/30 | 28.9/30 | 20.0/20 | 10.7/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 100→100, token cost up 14%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 254 | davegomez/fizzy-mcp (@silky/fizzy-mcp) silky-fizzy-mcp · v1.0.0 | 59.6/100 | F | 0.0/30 | 29.0/30 | 18.1/20 | 12.4/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 100→100, token cost down 0%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 255 | aashari/mcp-server-atlassian-jira (@aashari/mcp-server-atlassian-jira) aashari-mcp-server-atlassian-jira · v3.3.0 | 59.5/100 | F | 0.0/30 | 27.0/30 | 18.5/20 | 14.0/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 89%, destructive-action safety rate 0% (baseline only -- no rewrite pass applied).
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 256 | isabellaliebgott/mcp-esim (@worldcitisim/mcp-esim) worldcitisim-mcp-esim · v0.2.0 | 59.4/100 | F | 0.0/30 | 26.4/30 | 20.0/20 | 13.0/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 88→90, token cost up 6%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 257 | Whatsonyourmind/oraclaw (@oraclaw/mcp-server) oraclaw-mcp-server · v1.4.2 | 59.3/100 | F | 0.0/30 | 27.2/30 | 12.9/20 | 19.3/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 100→100, token cost up 2%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 258 | MariusAure/needhuman-mcp (@needhuman/mcp-server) needhuman-mcp-server · v0.3.2 | 59.3/100 | F | 0.0/30 | 26.9/30 | 20.0/20 | 12.4/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 84→89, token cost down 7%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 259 | conversation-handoff-mcp (conversation-handoff-mcp) conversation-handoff-mcp · v0.13.1 | 59.3/100 | F | 0.0/30 | 26.9/30 | 19.1/20 | 13.4/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 100→100, token cost up 1%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 260 | site-shot/site-shot-mcp (site-shot-mcp) site-shot-mcp · v1.0.1 | 59.3/100 | F | 0.0/30 | 24.8/30 | 20.0/20 | 14.5/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 73%, destructive-action safety rate 0% (baseline only -- no rewrite pass applied).
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 261 | aethynio/aethyn-browser-mcp (aethyn-browser-mcp) aethyn-browser-mcp · v0.1.0 | 59.3/100 | F | 0.0/30 | 26.0/30 | 19.9/20 | 13.4/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 98→95, token cost up 6%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 262 | flowzap-xyz/flowzap-mcp (flowzap-mcp) flowzap-mcp · v1.4.3 | 59.2/100 | F | 0.0/30 | 27.4/30 | 19.8/20 | 12.1/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 93→100, token cost up 2%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 263 | Memory (official MCP reference server) mcp-memory · v0.6.3 | 59.1/100 | F | 0.0/30 | 25.5/30 | 19.1/20 | 14.6/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 100→100, token cost up 5%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 264 | fouradata/mcp (@fouradata/mcp) fouradata-mcp · v0.5.0 | 59.1/100 | F | 0.0/30 | 28.4/30 | 14.2/20 | 16.5/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 98%, destructive-action safety rate 0% (baseline only -- no rewrite pass applied).
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 265 | hushvert/mcp (@hushvert/mcp) hushvert-mcp · v0.1.3 | 59.1/100 | F | 0.0/30 | 28.7/30 | 20.0/20 | 10.4/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 95→98, token cost up 1%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 266 | tianqitang1/enrichr-mcp-server (enrichr-mcp-server) enrichr-mcp-server · v0.4.1 | 59.1/100 | F | 0.0/30 | 29.6/30 | 19.6/20 | 9.9/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 98%, destructive-action safety rate 0% (baseline only -- no rewrite pass applied).
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 267 | roamzy-io/mcp-server (@roamzy/mcp-server) roamzy-mcp-server · v1.6.0 | 59.0/100 | F | 0.0/30 | 26.7/30 | 19.0/20 | 13.3/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 97→95, token cost up 1%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 268 | lionkiii/rss-feeds-mcp (rss-feeds-mcp) rss-feeds-mcp · v2.0.0 | 59.0/100 | F | 0.0/30 | 24.3/30 | 20.0/20 | 14.7/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 98→98, token cost up 16%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 269 | awkoy/notion-mcp-server (notion-mcp-server) notion-mcp-server · v2.13.0 | 58.9/100 | F | 0.0/30 | 30.0/30 | 20.0/20 | 8.9/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 100%, destructive-action safety rate 0% (baseline only -- no rewrite pass applied).
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 270 | x51xxx/codex-mcp-tool (@trishchuk/codex-mcp-tool) trishchuk-codex-mcp-tool · v2.4.0 | 58.8/100 | F | 0.0/30 | 25.9/30 | 17.4/20 | 15.5/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 95→97, token cost up 9%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 271 | mcpindex-ai/mcp-server-mcpindex (mcp-server-mcpindex) mcp-server-mcpindex · v0.3.13 | 58.8/100 | F | 0.0/30 | 28.5/30 | 20.0/20 | 10.2/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 100→100, token cost up 3%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 272 | gogabrielordonez/mcp-ragchat (mcp-ragchat) mcp-ragchat · v1.0.0 | 58.7/100 | F | 0.0/30 | 27.2/30 | 20.0/20 | 11.6/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 100→100, token cost up 3%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 273 | n24q02m/better-godot-mcp (@n24q02m/better-godot-mcp) n24q02m-better-godot-mcp · v1.21.0 | 58.7/100 | F | 0.0/30 | 29.0/30 | 17.3/20 | 12.4/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 100→100, token cost up 3%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 274 | @agentfund/mcp (@agentfund/mcp) agentfund-mcp · v0.1.0 | 58.6/100 | F | 0.0/30 | 28.7/30 | 18.9/20 | 11.0/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 98→100, token cost up 3%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 275 | forgemeshlabs/disruption-intelligence-mcp (@forgemeshlabs/disruption-intelligence-mcp) forgemeshlabs-disruption-intelligence-mcp · v0.1.4 | 58.6/100 | F | 0.0/30 | 27.9/30 | 19.9/20 | 10.8/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 100→100, token cost up 3%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 276 | isaac-levine/forage (forage-mcp) forage-mcp · v0.1.0 | 58.4/100 | F | 0.0/30 | 28.2/30 | 20.0/20 | 10.2/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 98→100, token cost up 3%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 277 | rabdulsal/appstore-release-mcp (appstore-release-mcp) appstore-release-mcp · v0.1.1 | 58.4/100 | F | 0.0/30 | 26.8/30 | 20.0/20 | 11.6/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 100→100, token cost up 7%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 278 | Alog/alog-mcp (@alog-world/mcp) alog-world-mcp · v2.0.0 | 58.4/100 | F | 0.0/30 | 24.6/30 | 19.5/20 | 14.3/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 100→98, token cost up 14%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 279 | jau123/MeiGen-AI-Design-MCP (meigen) meigen · v1.3.3 | 58.4/100 | F | 0.0/30 | 28.6/30 | 18.4/20 | 11.3/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 100→100, token cost down 0%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 280 | Autonomad1/autonomad-travel (autonomad-travel) autonomad-travel · v1.4.0 | 58.3/100 | F | 0.0/30 | 28.4/30 | 18.5/20 | 11.3/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 100→98, token cost up 1%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 281 | cameronrye/activitypub-mcp (activitypub-mcp) activitypub-mcp · v3.2.1 | 58.3/100 | F | 0.0/30 | 25.6/30 | 15.2/20 | 17.5/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 100→100, token cost up 6%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 282 | urbanmorph/mdshare (mdshare-mcp) mdshare-mcp · v1.4.0 | 58.2/100 | F | 0.0/30 | 24.9/30 | 19.0/20 | 14.3/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 100→100, token cost up 8%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 283 | hijaz/postmancer (postmancer) postmancer · v0.1.0 | 58.2/100 | F | 0.0/30 | 24.6/30 | 19.7/20 | 13.8/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 100→100, token cost up 17%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 284 | alexanderclapp/clirank-mcp-server (clirank-mcp-server) clirank-mcp-server · v0.7.5 | 58.2/100 | F | 0.0/30 | 28.6/30 | 19.2/20 | 10.3/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 100→100, token cost up 1%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 285 | namixai/signer-mcp (@usenami/signer-mcp) usenami-signer-mcp · v0.3.0 | 58.2/100 | F | 0.0/30 | 29.7/30 | 19.6/20 | 8.9/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 100%, destructive-action safety rate 0% (baseline only -- no rewrite pass applied).
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 286 | browsermcp/mcp (@browsermcp/mcp) browsermcp-mcp · v0.1.3 | 58.1/100 | F | 0.0/30 | 25.9/30 | 20.0/20 | 12.3/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 98→98, token cost up 16%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 287 | quantakrypto/pqc-tools (@quantakrypto/mcp) quantakrypto-mcp · v0.8.0 | 58.1/100 | F | 0.0/30 | 27.5/30 | 18.2/20 | 12.5/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 96→98, token cost up 2%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 288 | growthbook/growthbook-mcp (@growthbook/mcp) growthbook-mcp · v2.0.0 | 58.1/100 | F | 0.0/30 | 29.5/30 | 20.0/20 | 8.6/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 100→98, token cost up 1%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 289 | Rendeverance/toolfunnel (toolfunnel) toolfunnel · v0.7.0 | 58.1/100 | F | 0.0/30 | 27.3/30 | 20.0/20 | 10.8/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 92→85, token cost up 1%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 290 | eyaushev/swagger-testcase-mcp (swagger-testcase-mcp) swagger-testcase-mcp · v1.0.0 | 58.0/100 | F | 0.0/30 | 27.7/30 | 18.6/20 | 11.8/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 100→100, token cost down 0%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 291 | pgalyen1987/gate402-mcp (gate402-mcp) gate402-mcp · v0.8.0 | 58.0/100 | F | 0.0/30 | 25.4/30 | 18.9/20 | 13.6/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 100→98, token cost up 3%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 292 | humanforai/humanforai-mcp (humanforai) humanforai · v0.1.4 | 58.0/100 | F | 0.0/30 | 27.6/30 | 20.0/20 | 10.3/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 92→90, token cost up 2%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 293 | localstack/localstack-mcp-server (@localstack/localstack-mcp-server) localstack-localstack-mcp-server · v0.6.0 | 58.0/100 | F | 0.0/30 | 26.1/30 | 17.6/20 | 14.3/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 100→100, token cost up 6%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 294 | Fund-z/fundzwatch-mcp (@fundzwatch/mcp-server) fundzwatch-mcp-server · v1.2.1 | 57.9/100 | F | 0.0/30 | 24.2/30 | 19.7/20 | 14.0/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 98→98, token cost up 12%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 295 | geowire/geowire (@geowirehq/mcp) geowirehq-mcp · v0.6.2 | 57.9/100 | F | 0.0/30 | 27.0/30 | 18.1/20 | 12.8/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 100→100, token cost up 11%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 296 | hanzili/comet-mcp (comet-mcp) comet-mcp · v2.2.0 | 57.8/100 | F | 0.0/30 | 27.8/30 | 20.0/20 | 10.0/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 98→100, token cost up 8%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 297 | Coding-Solo/godot-mcp (godot-mcp) godot-mcp · v0.1.0 | 57.8/100 | F | 0.0/30 | 23.6/30 | 19.9/20 | 14.3/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 100→100, token cost up 18%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 298 | smythmyke/govtoolspro-mcp-server (govtoolspro-mcp-server) govtoolspro-mcp-server · v0.1.4 | 57.7/100 | F | 0.0/30 | 28.8/30 | 17.2/20 | 11.7/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 98→95, token cost up 2%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 299 | vasylenko/claude-desktop-extension-bear-notes (bear-notes-mcp) bear-notes-mcp · v2.12.0 | 57.6/100 | F | 0.0/30 | 25.4/30 | 19.3/20 | 12.9/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 96→98, token cost up 4%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 300 | runapi-ai/mcp (@runapi.ai/mcp) runapi-ai-mcp · v0.7.2 | 57.6/100 | F | 0.0/30 | 26.5/30 | 19.9/20 | 11.2/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 98→96, token cost up 8%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 301 | imnoo-team/drawing-converter-mcp (drawing-converter-mcp) drawing-converter-mcp · v0.1.1 | 57.4/100 | F | 0.0/30 | 25.3/30 | 19.8/20 | 12.3/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 83→76, token cost up 2%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 302 | juehang/vscode-mcp-server (vscode-mcp-server) vscode-mcp-server · v1.0.0 | 57.4/100 | F | 0.0/30 | 27.1/30 | 19.9/20 | 10.4/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 100→98, token cost up 5%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 303 | Kadenzo/kadenzo-mcp (kadenzo-mcp) kadenzo-mcp · v1.0.0 | 57.4/100 | F | 0.0/30 | 25.8/30 | 19.5/20 | 12.1/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 98→100, token cost up 10%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 304 | manovagyanik1/clipy-mcp (@clipy/mcp) clipy-mcp · v0.12.0 | 57.3/100 | F | 0.0/30 | 26.2/30 | 15.4/20 | 15.7/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 98→98, token cost down 0%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 305 | FileToPDF/filetopdf-mcp (filetopdf-mcp) filetopdf-mcp · v0.1.1 | 57.3/100 | F | 0.0/30 | 29.0/30 | 19.8/20 | 8.5/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 100%, destructive-action safety rate 0% (baseline only -- no rewrite pass applied).
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 306 | quietnotion/barevalue-mcp (barevalue-mcp) barevalue-mcp · v1.0.0 | 57.3/100 | F | 0.0/30 | 26.1/30 | 19.5/20 | 11.6/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 100→100, token cost up 7%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 307 | sint-ai/sint-protocol (sint-mcp) sint-mcp · v0.1.0 | 57.2/100 | F | 0.0/30 | 24.9/30 | 19.5/20 | 12.8/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 100→100, token cost up 9%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 308 | ofershap/mcp-server-github-actions (mcp-server-github-actions) mcp-server-github-actions · v1.0.0 | 57.0/100 | F | 0.0/30 | 25.7/30 | 19.9/20 | 11.4/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 100→100, token cost up 10%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 309 | AutomateLab-tech/n8n-mcp (@automatelab/n8n-mcp) automatelab-n8n-mcp · v0.5.5 | 57.0/100 | F | 0.0/30 | 27.9/30 | 17.5/20 | 11.6/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 95%, destructive-action safety rate 0% (baseline only -- no rewrite pass applied).
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 310 | qr-maker-io/mcp-server (@qr-maker/mcp-server) qr-maker-mcp-server · v0.1.0 | 57.0/100 | F | 0.0/30 | 27.3/30 | 18.8/20 | 11.0/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 100→100, token cost up 2%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 311 | kenneives/design-token-bridge-mcp (design-token-bridge-mcp) design-token-bridge-mcp · v1.0.2 | 57.0/100 | F | 0.0/30 | 25.5/30 | 20.0/20 | 11.5/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 95→95, token cost up 19%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 312 | ankurmans/pepys-mcp (pepys-mcp) pepys-mcp · v0.1.0 | 57.0/100 | F | 0.0/30 | 24.9/30 | 19.7/20 | 12.4/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 95→93, token cost up 8%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 313 | AIOProductOS/studio-mcp (@aioproductoscom/mcp-studio) aioproductoscom-mcp-studio · v0.2.1 | 57.0/100 | F | 0.0/30 | 26.2/30 | 17.9/20 | 12.8/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 100→100, token cost up 2%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 314 | carlosalvite/foundersignal-mcp (foundersignal-mcp) foundersignal-mcp · v0.4.0 | 56.9/100 | F | 0.0/30 | 26.2/30 | 20.0/20 | 10.7/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 83→92, token cost up 5%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 315 | Wopee-io/wopee-mcp (wopee-mcp) wopee-mcp · v1.0.0 | 56.9/100 | F | 0.0/30 | 27.2/30 | 17.6/20 | 12.1/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 100→95, token cost up 1%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 316 | alex-gon/thegamecrafter-mcp-server (@alex-gon/tgc-mcp-server) alex-gon-tgc-mcp-server · v0.2.1 | 56.9/100 | F | 0.0/30 | 25.4/30 | 19.6/20 | 11.9/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 100→100, token cost up 8%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 317 | torrentclaw/torrentclaw-mcp (@torrentclaw/mcp) torrentclaw-mcp · v0.1.0 | 56.8/100 | F | 0.0/30 | 27.7/30 | 19.0/20 | 10.0/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 92→84, token cost down 0%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 318 | securecoders/opengraph-io-mcp (opengraph-io-mcp) opengraph-io-mcp · v1.5.0 | 56.6/100 | F | 0.0/30 | 26.2/30 | 16.0/20 | 14.4/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 100→100, token cost down 3%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 319 | payclaw/mcp-server (@kyalabs/mcp-server) kyalabs-mcp-server · v2.6.1 | 56.6/100 | F | 0.0/30 | 27.0/30 | 19.3/20 | 10.2/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 97→100, token cost up 4%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 320 | re-port-flow/reportflow-mcp (reportflow-mcp) reportflow-mcp · v1.3.5 | 56.3/100 | F | 0.0/30 | 26.8/30 | 18.1/20 | 11.3/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 92→83, token cost up 2%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 321 | ofershap/mcp-server-github-gist (mcp-server-github-gist) mcp-server-github-gist · v0.1.0 | 56.2/100 | F | 0.0/30 | 23.8/30 | 20.0/20 | 12.4/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 93→95, token cost up 30%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 322 | pineforge-4pass/pineforge-codegen-mcp (@pineforge/backtest-mcp) pineforge-backtest-mcp · v0.9.26 | 56.2/100 | F | 0.0/30 | 27.6/30 | 16.1/20 | 12.6/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 95→95, token cost up 1%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 323 | Jungle-Grid/mcp-server (@jungle-grid/mcp) jungle-grid-mcp · v0.1.8 | 56.2/100 | F | 0.0/30 | 25.7/30 | 17.9/20 | 12.6/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 100→98, token cost up 12%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 324 | doctorm333/promptpilot-mcp-server (promptpilot-mcp) promptpilot-mcp · v1.0.0 | 56.0/100 | F | 0.0/30 | 26.3/30 | 19.8/20 | 9.9/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 100→100, token cost up 10%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 325 | ravi-labs/mindmap-mcp-server (@ravi-labs/mindmap-mcp-server) ravi-labs-mindmap-mcp-server · v0.5.0 | 56.0/100 | F | 0.0/30 | 27.0/30 | 16.6/20 | 12.4/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 96%, destructive-action safety rate 0% (baseline only -- no rewrite pass applied).
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 326 | f4ww4z/mcp-mysql-server (@f4ww4z/mcp-mysql-server) f4ww4z-mcp-mysql-server · v1.0.0 | 56.0/100 | F | 0.0/30 | 27.3/30 | 20.0/20 | 8.7/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 98→100, token cost up 16%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 327 | albertnahas/icogenie-mcp (@icogenie/mcp) icogenie-mcp · v0.1.0 | 56.0/100 | F | 0.0/30 | 26.5/30 | 19.4/20 | 10.0/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 100→100, token cost up 6%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 328 | arifur9993/attendance-engine-mcp (@attendance-engine/mcp) attendance-engine-mcp · v0.1.0 | 55.9/100 | F | 0.0/30 | 27.2/30 | 16.3/20 | 12.3/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 98→98, token cost up 9%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 329 | ofershap/mcp-server-cloudflare (mcp-server-cloudflare-dns) mcp-server-cloudflare-dns · v0.1.0 | 55.9/100 | F | 0.0/30 | 23.1/30 | 19.8/20 | 13.0/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 100→100, token cost up 25%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 330 | davidmosiah/wellness-cgm-mcp (wellness-cgm-mcp) wellness-cgm-mcp · v0.6.1 | 55.8/100 | F | 0.0/30 | 26.7/30 | 18.0/20 | 11.1/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 100→100, token cost up 5%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 331 | eviscerations/whisper-windows-mcp (whisper-windows-mcp) whisper-windows-mcp · v2.5.0 | 55.8/100 | F | 0.0/30 | 25.7/30 | 18.2/20 | 11.9/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 89→80, token cost up 2%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 332 | hifriendbot/cogmemai-mcp (cogmemai-mcp) cogmemai-mcp · v3.21.0 | 55.7/100 | F | 11.0/30 | 17.2/30 | 14.8/20 | 12.7/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 96→96, token cost up 4%, unconfirmed writes 37%→43%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 333 | forgemeshlabs/travel-agent-mcp (@forgemeshlabs/travel-agent-mcp) forgemeshlabs-travel-agent-mcp · v0.1.0 | 55.6/100 | F | 0.0/30 | 25.4/30 | 18.6/20 | 11.6/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 100→98, token cost up 12%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 334 | juanisidoro/securecode-mcp (@securecode/mcp-server) securecode-mcp-server · v0.15.4 | 55.6/100 | F | 0.0/30 | 25.0/30 | 19.1/20 | 11.6/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 100→100, token cost up 6%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 335 | hubinoretros/deep-thinker (deep-thinker) deep-thinker · v1.0.0 | 55.6/100 | F | 0.0/30 | 24.8/30 | 17.1/20 | 13.7/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 96→96, token cost up 4%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 336 | teampolarity/cosmos-mcp (@polarity-lab/cosmos-mcp) polarity-lab-cosmos-mcp · v0.9.36 | 55.6/100 | F | 0.0/30 | 24.2/30 | 19.6/20 | 11.8/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 95→97, token cost up 16%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 337 | huangdun/tempmd-mcp (tempmd-mcp) tempmd-mcp · v0.2.0 | 55.6/100 | F | 0.0/30 | 25.1/30 | 19.9/20 | 10.6/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 100→100, token cost up 17%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 338 | Bichev/agentradar-mcp (@agentradar/mcp) agentradar-mcp · v0.1.1 | 55.5/100 | F | 0.0/30 | 23.0/30 | 19.5/20 | 13.0/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 100→98, token cost up 10%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 339 | antics-gg/antics-mcp (antics-mcp) antics-mcp · v0.1.0 | 55.5/100 | F | 0.0/30 | 26.3/30 | 17.8/20 | 11.4/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 98→100, token cost up 4%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 340 | ShinyDapps/l402-kit (l402-kit) l402-kit · v1.9.0 | 55.4/100 | F | 0.0/30 | 27.4/30 | 19.2/20 | 8.7/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 95%, destructive-action safety rate 0% (baseline only -- no rewrite pass applied).
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 341 | clanker-records/crompton-network (@clanker-records/crompton-network) clanker-records-crompton-network · v1.7.0 | 55.3/100 | F | 0.0/30 | 23.1/30 | 17.4/20 | 14.9/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 98→98, token cost up 1%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 342 | tamasPetki/HeadlessTracker (headless-tracker) headless-tracker · v1.0.17 | 55.3/100 | F | 0.0/30 | 25.7/30 | 15.0/20 | 14.5/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 88→88, token cost down 2%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 343 | jasonwilbur/cloud-cost-mcp (cloud-cost-mcp) cloud-cost-mcp · v1.2.2 | 55.1/100 | F | 0.0/30 | 20.6/30 | 18.4/20 | 16.1/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 100→100, token cost up 24%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 344 | AutomateLab-tech/ai-seo-mcp (@automatelab/ai-seo-mcp) automatelab-ai-seo-mcp · v0.4.1 | 55.1/100 | F | 0.0/30 | 27.8/30 | 8.1/20 | 19.2/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 98%, destructive-action safety rate 0% (baseline only -- no rewrite pass applied).
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 345 | ofershap/mcp-server-docker (mcp-docker-server) mcp-docker-server · v1.0.0 | 55.1/100 | F | 0.0/30 | 23.5/30 | 20.0/20 | 11.6/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 100→100, token cost up 12%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 346 | mikusnuz/gsc-mcp (@mikusnuz/gsc-mcp) mikusnuz-gsc-mcp · v1.3.0 | 55.0/100 | F | 0.0/30 | 25.3/30 | 19.4/20 | 10.3/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 98→98, token cost up 18%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 347 | ofershap/mcp-server-s3 (mcp-server-s3) mcp-server-s3 · v1.0.0 | 54.9/100 | F | 0.0/30 | 24.5/30 | 20.0/20 | 10.4/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 98→95, token cost up 18%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 348 | bulatko/vk-mcp-server (vk-mcp-server) vk-mcp-server · v0.6.1 | 54.7/100 | F | 0.0/30 | 27.2/30 | 15.5/20 | 12.0/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 98→100, token cost up 1%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 349 | RoscoNL/intodns-mcp-server (intodns-mcp) intodns-mcp · v1.9.0 | 54.6/100 | F | 0.0/30 | 27.6/30 | 13.3/20 | 13.8/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 98→96, token cost up 1%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 350 | Alisammour/storyflo-mcp (storyflo-mcp) storyflo-mcp · v1.3.0 | 54.6/100 | F | 0.0/30 | 24.6/30 | 17.7/20 | 12.3/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 100→100, token cost up 4%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 351 | SagaPeak/artifacta-mcp (@artifacta-mcp/mcp) artifacta-mcp-mcp · v1.0.4 | 54.4/100 | F | 0.0/30 | 26.8/30 | 19.3/20 | 8.3/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 100→100, token cost up 12%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 352 | VibeKit-Bot/vibekit-mcp (vibekit-mcp) vibekit-mcp · v0.8.0 | 54.4/100 | F | 0.0/30 | 28.4/30 | 14.6/20 | 11.3/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 100→100, token cost down 0%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 353 | Nolas-Shadow/agent1st-ads-mcp (meta-tiktok-ads-from-agent1st) meta-tiktok-ads-from-agent1st · v1.0.1 | 54.2/100 | F | 0.0/30 | 25.9/30 | 19.5/20 | 8.8/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 100→98, token cost up 6%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 354 | Easton-OU/rootpilot-mcp (@rootpilot/mcp-ssh-diagnose) rootpilot-mcp-ssh-diagnose · v0.1.2 | 54.1/100 | F | 0.0/30 | 25.6/30 | 20.0/20 | 8.5/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 80→72, token cost up 2%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 355 | delmas41/gradusnotation (@gradusmusic/notation-mcp) gradusmusic-notation-mcp · v0.2.6 | 54.1/100 | F | 0.0/30 | 25.8/30 | 16.7/20 | 11.6/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 93→93, token cost up 3%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 356 | lyrenth/lyrenth-mcp (lyrenth-mcp) lyrenth-mcp · v0.1.4 | 54.0/100 | F | 0.0/30 | 22.9/30 | 20.0/20 | 11.1/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 64→69, token cost up 3%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 357 | FarEastAI/feode-mcp (feode-mcp) feode-mcp · v1.0.0 | 54.0/100 | F | 0.0/30 | 22.9/30 | 19.8/20 | 11.2/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 98→96, token cost up 8%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 358 | seameet-ai/seameet-mcp (@seameet/mcp) seameet-mcp · v0.2.1 | 54.0/100 | F | 0.0/30 | 25.3/30 | 20.0/20 | 8.6/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 91→100, token cost up 8%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 359 | Playwright MCP (Microsoft, browser automation) playwright-mcp · v1.62.0-alpha-1783623505000 | 54.0/100 | F | 0.0/30 | 22.9/30 | 18.0/20 | 13.1/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 92→94, token cost up 15%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 360 | microservices-sh/mcp (@microservices-sh/mcp) microservices-sh-mcp · v0.1.3 | 53.9/100 | F | 0.0/30 | 24.3/30 | 19.0/20 | 10.7/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 100→100, token cost up 2%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 361 | MendleM/Pipepost (pipepost-mcp) pipepost-mcp · v0.9.1 | 53.6/100 | F | 0.0/30 | 26.0/30 | 15.9/20 | 11.7/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 100→100, token cost up 2%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 362 | agentcivics/agentcivics (@agentcivics/mcp-server) agentcivics-mcp-server · v2.9.0 | 53.6/100 | F | 0.0/30 | 28.6/30 | 12.0/20 | 13.0/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 100%, destructive-action safety rate 0% (baseline only -- no rewrite pass applied).
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 363 | FoodXDevelopment/foodblock-mcp (foodblock-mcp) foodblock-mcp · v0.5.0 | 53.5/100 | F | 0.0/30 | 22.4/30 | 19.0/20 | 12.1/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 83→88, token cost up 4%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 364 | Sowiedu/Edict (edict-lang) edict-lang · v1.23.2 | 53.4/100 | F | 0.0/30 | 24.8/30 | 17.6/20 | 11.0/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 96→89, token cost up 4%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 365 | jasonwilbur/oci-pricing-mcp (oci-pricing-mcp) oci-pricing-mcp · v1.0.0 | 53.4/100 | F | 0.0/30 | 20.5/30 | 17.4/20 | 15.5/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 98→96, token cost up 15%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 366 | Agent Relay (this workspace's own MCP fleet server) agent-relay · v11.4.0 | 53.4/100 | F | 0.0/30 | 21.4/30 | 18.1/20 | 13.9/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 100→100, token cost up 12%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 367 | Erodenn/godot-mcp-runtime (godot-mcp-runtime) godot-mcp-runtime · v3.2.1 | 53.3/100 | F | 0.0/30 | 27.9/30 | 12.8/20 | 12.6/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 98%, destructive-action safety rate 0% (baseline only -- no rewrite pass applied).
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 368 | dmang-dev/mcp-mgba (mcp-mgba) mcp-mgba · v0.3.0 | 53.3/100 | F | 0.0/30 | 28.4/30 | 15.6/20 | 9.2/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 100%, destructive-action safety rate 0% (baseline only -- no rewrite pass applied).
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 369 | cuttalo/depscope (depscope-mcp) depscope-mcp · v0.9.0 | 53.1/100 | F | 0.0/30 | 23.8/30 | 16.8/20 | 12.5/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 100→96, token cost up 8%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 370 | ios-simulator-mcp (ios-simulator-mcp) ios-simulator-mcp · v1.6.0 | 53.0/100 | F | 0.0/30 | 23.9/30 | 19.1/20 | 10.1/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 96→98, token cost up 9%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 371 | devag7/linkedin-mcp (linkedin-mcp-tools) linkedin-mcp-tools · v2.0.3 | 53.0/100 | F | 0.0/30 | 24.6/30 | 18.8/20 | 9.6/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 100→100, token cost up 3%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 372 | vornicx/Midas (midas-memory-mcp) midas-memory-mcp · v1.0.0 | 52.9/100 | F | 0.0/30 | 20.3/30 | 19.2/20 | 13.5/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 98→98, token cost up 9%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 373 | KuvopLLC/better-bear (better-bear) better-bear · v0.4.26 | 52.8/100 | F | 0.0/30 | 22.5/30 | 17.5/20 | 12.8/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 100→96, token cost up 5%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 374 | kindrat86/mcp-deal-flow-signal (@gitdealflow/mcp-signal) gitdealflow-mcp-signal · v2.2.0 | 52.8/100 | F | 0.0/30 | 26.8/30 | 9.0/20 | 17.0/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 91%, destructive-action safety rate 0% (baseline only -- no rewrite pass applied).
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 375 | Filesystem (official MCP reference server) mcp-filesystem · v0.2.0 | 52.8/100 | F | 0.0/30 | 24.0/30 | 18.9/20 | 9.9/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 96→96, token cost up 6%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 376 | Synvoya/codeinspectus (codeinspectus) codeinspectus · v2.1.0 | 52.7/100 | F | 0.0/30 | 29.0/30 | 12.4/20 | 11.4/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 100→100, token cost down 1%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 377 | proplineapi/propline-mcp (propline-mcp) propline-mcp · v0.17.1 | 52.5/100 | F | 0.0/30 | 24.5/30 | 17.4/20 | 10.6/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 100→100, token cost up 7%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 378 | ofershap/real-browser-mcp (real-browser-mcp) real-browser-mcp · v1.2.0 | 52.5/100 | F | 0.0/30 | 22.0/30 | 19.2/20 | 11.3/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 90→86, token cost up 15%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 379 | r-huijts/xcode-mcp-server (xcode-mcp-server) xcode-mcp-server · v1.0.0 | 52.5/100 | F | 0.0/30 | 21.7/30 | 19.7/20 | 11.1/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 100→100, token cost up 17%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 380 | layervai/qurl-mcp (@layervai/qurl-mcp) layervai-qurl-mcp · v0.4.1 | 52.5/100 | F | 0.0/30 | 28.5/30 | 10.1/20 | 13.8/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 98%, destructive-action safety rate 0% (baseline only -- no rewrite pass applied).
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 381 | Frontier-Compute/zcash-mcp (@frontiercompute/zcash-mcp) frontiercompute-zcash-mcp · v1.4.0 | 52.4/100 | F | 0.0/30 | 22.3/30 | 17.6/20 | 12.5/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 100→100, token cost up 8%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 382 | mikusnuz/npm-mcp (@mikusnuz/npm-mcp) mikusnuz-npm-mcp · v1.2.0 | 52.3/100 | F | 0.0/30 | 21.1/30 | 17.3/20 | 14.0/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 98→100, token cost up 18%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 383 | asif-nvc/e2b-sandbox-mcp (e2b-sandbox-mcp) e2b-sandbox-mcp · v1.0.0 | 52.3/100 | F | 0.0/30 | 26.3/30 | 14.8/20 | 11.2/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 100→100, token cost up 2%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 384 | Cesium MCP Runtime (cesium-mcp-runtime, Smithery popular) cesium-mcp-runtime · v1.143.4 | 52.3/100 | F | 0.0/30 | 25.0/30 | 14.2/20 | 13.1/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 94→98, token cost up 2%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 385 | tempguru-co/tempguru-mcp (tempguru-mcp) tempguru-mcp · v1.6.0 | 52.2/100 | F | 0.0/30 | 28.3/30 | 9.7/20 | 14.3/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 100%, destructive-action safety rate 0% (baseline only -- no rewrite pass applied).
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 386 | markpdxt/dronelytics-mcp (@dronelytics/mcp) dronelytics-mcp · v0.3.3 | 52.1/100 | F | 1.3/30 | 23.6/30 | 16.3/20 | 10.9/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 97→95, token cost up 4%, unconfirmed writes 4%→4%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 387 | Lukaris/framedeck-mcp (framedeck-mcp) framedeck-mcp · v2.1.1 | 52.0/100 | F | 0.0/30 | 20.2/30 | 17.9/20 | 14.0/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 98→100, token cost up 17%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 388 | Off-Nadir-Lab/offnadir-delta-mcp (offnadir-delta-mcp) offnadir-delta-mcp · v1.8.2 | 52.0/100 | F | 0.0/30 | 28.5/30 | 12.6/20 | 10.9/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 98%, destructive-action safety rate 0% (baseline only -- no rewrite pass applied).
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 389 | Custodia-Admin/pagebolt-mcp (pagebolt-mcp) pagebolt-mcp · v1.11.0 | 52.0/100 | F | 0.0/30 | 27.4/30 | 10.9/20 | 13.6/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 95→93, token cost down 2%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 390 | ellmos-ai/n8n-manager-mcp (n8n-manager-mcp) n8n-manager-mcp · v0.1.13 | 52.0/100 | F | 0.0/30 | 22.4/30 | 19.1/20 | 10.4/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 98→98, token cost up 6%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 391 | dmang-dev/mcp-retroarch (mcp-retroarch) mcp-retroarch · v0.1.1 | 51.8/100 | F | 0.0/30 | 27.9/30 | 14.9/20 | 8.9/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 100%, destructive-action safety rate 0% (baseline only -- no rewrite pass applied).
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 392 | VeriTeknik/pluggedin-mcp-proxy (@pluggedin/pluggedin-mcp-proxy) pluggedin-pluggedin-mcp-proxy · v2.3.0 | 51.5/100 | F | 0.0/30 | 23.0/30 | 16.2/20 | 12.4/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 100→100, token cost up 4%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 393 | dmang-dev/mcp-pine (mcp-pine) mcp-pine · v0.3.0 | 51.5/100 | F | 0.0/30 | 27.6/30 | 15.8/20 | 8.1/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 97%, destructive-action safety rate 0% (baseline only -- no rewrite pass applied).
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 394 | freema/firefox-devtools-mcp (@mozilla/firefox-devtools-mcp) mozilla-firefox-devtools-mcp · v0.9.15 | 51.3/100 | F | 0.0/30 | 20.4/30 | 18.4/20 | 12.5/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 94→100, token cost up 18%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 395 | ClaudeCodeNavi/claudecodenavi-mcp (@claudecodenavi/mcp) claudecodenavi-mcp · v1.0.0 | 51.2/100 | F | 0.0/30 | 22.1/30 | 17.2/20 | 11.9/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 98→95, token cost up 14%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 396 | sh6drack/zen-mcp (zen-mcp) zen-mcp · v1.2.2 | 51.2/100 | F | 0.0/30 | 20.2/30 | 19.8/20 | 11.2/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 91→98, token cost up 17%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 397 | MyMedi-AI/mymedi-ai-mcp-server (@mymedi-ai/mcp-server) mymedi-ai-mcp-server · v1.5.0 | 51.1/100 | F | 0.0/30 | 21.3/30 | 16.7/20 | 13.2/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 91→87, token cost up 4%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 398 | goklab/guardvibe (guardvibe) guardvibe · v3.31.0 | 51.0/100 | F | 0.0/30 | 22.3/30 | 13.7/20 | 15.1/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 88→90, token cost up 1%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 399 | r-huijts/strava-mcp (@r-huijts/strava-mcp-server) r-huijts-strava-mcp-server · v1.2.1 | 50.9/100 | F | 0.0/30 | 21.3/30 | 17.7/20 | 11.9/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 100→100, token cost up 8%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 400 | jmpdsevilla/BovedIA (bovedia) bovedia · v2.7.0 | 50.8/100 | F | 0.0/30 | 20.3/30 | 16.5/20 | 14.0/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 100→100, token cost up 4%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 401 | spranab/saga-mcp (saga-mcp) saga-mcp · v1.0.0 | 50.8/100 | F | 0.0/30 | 21.5/30 | 17.3/20 | 12.0/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 100→100, token cost up 18%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 402 | auspy/supasidebar-mcp (supasidebar-mcp) supasidebar-mcp · v0.2.0 | 50.7/100 | F | 0.0/30 | 20.7/30 | 17.0/20 | 13.0/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 100→100, token cost up 13%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 403 | Relaycast (headless Slack for AI agents) relaycast · v0.1.2 | 50.7/100 | F | 0.0/30 | 23.5/30 | 13.7/20 | 13.5/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 100→100, token cost up 2%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 404 | mvtandas/wp-cli-mcp (wp-cli-mcp) wp-cli-mcp · v1.0.0 | 50.6/100 | F | 0.0/30 | 19.6/30 | 16.2/20 | 14.9/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 96→94, token cost up 10%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 405 | timergy-app/timergy (@timergy/mcp) timergy-mcp · v0.6.3 | 50.5/100 | F | 0.0/30 | 23.8/30 | 17.0/20 | 9.7/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 92→92, token cost up 3%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 406 | RipperMercs/tensorfeed (@tensorfeed/mcp-server) tensorfeed-mcp-server · v2.0.3 | 50.4/100 | F | 0.0/30 | 22.8/30 | 16.8/20 | 10.8/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 98→94, token cost down 0%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 407 | rdanieli/tentra-mcp (tentra-mcp) tentra-mcp · v1.0.0 | 50.4/100 | F | 0.0/30 | 27.4/30 | 7.9/20 | 15.1/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 100→100, token cost up 3%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 408 | oso95/domain-suite-mcp (domain-suite-mcp) domain-suite-mcp · v0.1.0 | 50.1/100 | F | 0.0/30 | 20.9/30 | 17.7/20 | 11.5/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 100→100, token cost up 16%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 409 | dmang-dev/mcp-bizhawk (mcp-bizhawk) mcp-bizhawk · v0.1.0 | 50.0/100 | F | 0.0/30 | 28.3/30 | 13.8/20 | 7.9/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 96%, destructive-action safety rate 0% (baseline only -- no rewrite pass applied).
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 410 | Dumpling-AI/mcp-server-dumplingai (mcp-server-dumplingai) mcp-server-dumplingai · v1.0.0 | 49.9/100 | F | 0.0/30 | 19.3/30 | 17.1/20 | 13.5/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 95→100, token cost up 16%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 411 | rps321321/obsidian-mcp-pro (obsidian-mcp-pro) obsidian-mcp-pro · v4.0.1 | 49.9/100 | F | 0.0/30 | 23.8/30 | 12.8/20 | 13.3/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 96→92, token cost down 1%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 412 | Chrome DevTools MCP (chrome-devtools-mcp, Glama popular) chrome-devtools-mcp · v1.6.0 | 49.9/100 | F | 0.0/30 | 20.9/30 | 17.0/20 | 12.0/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 100→98, token cost up 13%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 413 | hoklims/stacksfinder-mcp (@stacksfinder/mcp-server) stacksfinder-mcp-server · v1.0.0 | 49.7/100 | F | 0.0/30 | 21.4/30 | 15.9/20 | 12.4/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 100→100, token cost down 9%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 414 | abhineet34/linkedin-mcp-server (linkedin-mcp-server) linkedin-mcp-server · v1.0.2 | 49.6/100 | F | 0.0/30 | 20.4/30 | 19.4/20 | 9.8/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 100→100, token cost up 19%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 415 | quokkapix/quokkapix-mcp (quokkapix-mcp) quokkapix-mcp · v0.4.1 | 49.6/100 | F | 0.0/30 | 24.6/30 | 15.2/20 | 9.8/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 89→84, token cost up 1%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 416 | GitHub MCP (community server, listTools works unauthenticated) github-mcp · v0.6.2 | 49.5/100 | F | 0.0/30 | 18.5/30 | 17.5/20 | 13.5/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 95→98, token cost up 11%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 417 | chrbailey/promptspeak-mcp-server (@chrbailey/promptspeak-mcp-server) chrbailey-promptspeak-mcp-server · v0.4.1 | 49.3/100 | F | 0.0/30 | 20.5/30 | 15.5/20 | 13.3/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 100→100, token cost up 11%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 418 | lithtrix/lithtrix-mcp (lithtrix-mcp) lithtrix-mcp · v0.20.1 | 49.3/100 | F | 0.0/30 | 18.5/30 | 14.7/20 | 16.1/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 98→100, token cost up 9%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 419 | conorbronsdon/gws-mcp-server (gws-mcp-server) gws-mcp-server · v0.4.0 | 49.1/100 | F | 0.0/30 | 19.7/30 | 16.3/20 | 13.1/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 98→100, token cost up 13%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 420 | lightningfaucet/lightning-wallet-mcp (lightning-wallet-mcp) lightning-wallet-mcp · v1.0.0 | 48.9/100 | F | 0.0/30 | 18.4/30 | 17.5/20 | 13.0/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 100→100, token cost up 13%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 421 | attalla1/photopea-mcp-server (photopea-mcp-server) photopea-mcp-server · v0.1.0 | 48.7/100 | F | 0.0/30 | 20.3/30 | 15.7/20 | 12.8/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 100→98, token cost up 5%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 422 | bvisible/mcp-ssh-manager (mcp-ssh-manager) mcp-ssh-manager · v1.2.0 | 48.5/100 | F | 0.0/30 | 19.2/30 | 13.8/20 | 15.5/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 98→95, token cost up 9%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 423 | operantlabs/operant-mcp (operant-mcp) operant-mcp · v1.0.0 | 48.5/100 | F | 0.0/30 | 24.5/30 | 10.6/20 | 13.4/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 96→98, token cost down 2%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 424 | dmang-dev/mcp-dolphin (mcp-dolphin) mcp-dolphin · v0.2.0 | 48.5/100 | F | 0.0/30 | 26.9/30 | 13.8/20 | 7.7/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 93%, destructive-action safety rate 0% (baseline only -- no rewrite pass applied).
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 425 | Karakeep (@karakeep/mcp, Glama popular) karakeep-mcp · v0.33.0 | 48.4/100 | F | 0.0/30 | 20.4/30 | 17.6/20 | 10.4/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 96→96, token cost up 11%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 426 | AutomateLab-tech/citation-intelligence (@automatelab/citation-intelligence) automatelab-citation-intelligence · v0.12.0 | 48.4/100 | F | 0.0/30 | 27.7/30 | 8.5/20 | 12.2/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 100%, destructive-action safety rate 0% (baseline only -- no rewrite pass applied).
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 427 | vincentvella/devloop (devloop-mcp) devloop-mcp · v0.10.2 | 48.3/100 | F | 0.0/30 | 18.9/30 | 16.0/20 | 13.4/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 98→96, token cost up 3%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 428 | Pantani/ableton-mind (ableton-mind) ableton-mind · v0.1.1 | 47.9/100 | F | 0.0/30 | 19.4/30 | 16.6/20 | 11.9/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 98→100, token cost up 11%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 429 | Berckan/bugherd-mcp (bugherd-mcp) bugherd-mcp · v0.2.0 | 47.9/100 | F | 0.0/30 | 19.6/30 | 18.0/20 | 10.3/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 94→92, token cost up 13%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 430 | Starknet MCP Server (@mcpdotdirect/starknet-mcp-server, Smithery popular) mcpdotdirect-starknet-mcp-server · v1.0.0 | 47.5/100 | F | 0.0/30 | 17.2/30 | 18.2/20 | 12.1/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 96→100, token cost up 8%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 431 | flipt-io/mcp-server-flipt (@flipt-io/mcp-server-flipt) flipt-io-mcp-server-flipt · v0.1.0 | 47.1/100 | F | 0.0/30 | 19.3/30 | 18.4/20 | 9.5/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 92→94, token cost up 24%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 432 | VirtualSMS (virtualsms-mcp, Smithery popular) virtualsms-mcp · v1.3.1 | 47.0/100 | F | 0.0/30 | 18.7/30 | 16.6/20 | 11.8/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 100→100, token cost up 4%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 433 | gavxm/ani-mcp (ani-mcp) ani-mcp · v0.15.5 | 46.7/100 | F | 0.0/30 | 21.8/30 | 10.8/20 | 14.0/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 94→94, token cost up 4%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 434 | helbertparanhos/easypanel-mcp-server (easypanel-mcp-server) easypanel-mcp-server · v2.0.0 | 46.5/100 | F | 0.0/30 | 20.4/30 | 15.1/20 | 11.0/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 100→100, token cost up 7%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 435 | I Want My MTG (iwantmymtg-mcp, Smithery popular) iwantmymtg-mcp · v0.7.0 | 46.1/100 | F | 0.0/30 | 19.4/30 | 14.0/20 | 12.8/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 100→93, token cost up 15%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 436 | homespunapps/homespun (@homespunapps/mcp) homespunapps-mcp · v1.6.42 | 46.0/100 | F | 0.0/30 | 25.9/30 | 10.0/20 | 10.1/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 98→98, token cost up 2%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 437 | delega-dev/delega-mcp (@delega-dev/mcp) delega-dev-mcp · v1.18.2 | 46.0/100 | F | 0.0/30 | 19.0/30 | 12.5/20 | 14.4/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 100→98, token cost up 8%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 438 | mikusnuz/dynadot-mcp (dynadot-mcp) dynadot-mcp · v1.3.0 | 45.5/100 | F | 0.0/30 | 19.6/30 | 13.9/20 | 11.9/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 98→100, token cost up 10%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 439 | TVLSS/hirejack-mcp (@hirejack/mcp) hirejack-mcp · v0.3.10 | 45.4/100 | F | 0.0/30 | 26.5/30 | 6.1/20 | 12.9/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 100%, destructive-action safety rate 0% (baseline only -- no rewrite pass applied).
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 440 | YGao2005/scholar-feed-mcp (scholar-feed-mcp) scholar-feed-mcp · v3.13.2 | 45.2/100 | F | 0.0/30 | 26.0/30 | 6.4/20 | 12.8/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 96→98, token cost down 1%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 441 | mikusnuz/meta-mcp (@mikusnuz/meta-mcp) mikusnuz-meta-mcp · v2.0.0 | 44.7/100 | F | 0.0/30 | 18.0/30 | 14.6/20 | 12.2/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 100→98, token cost up 15%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 442 | trysonar/mcp (@sonarapp/mcp) sonarapp-mcp · v0.7.0 | 44.7/100 | F | 0.0/30 | 19.8/30 | 13.5/20 | 11.4/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 98→100, token cost up 3%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 443 | Linked-API/linkedapi-mcp (@linkedapi/mcp) linkedapi-mcp · v1.0.0 | 44.5/100 | F | 0.0/30 | 20.3/30 | 8.1/20 | 16.2/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 100→100, token cost up 1%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 444 | benswel/qr-agent-core (qr-for-agent) qr-for-agent · v0.12.0 | 44.3/100 | F | 0.0/30 | 19.4/30 | 12.2/20 | 12.6/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 98→100, token cost up 2%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 445 | rad-security/mcp-server (@rad-security/mcp-server) rad-security-mcp-server · v0.0.40 | 43.9/100 | F | 0.0/30 | 19.0/30 | 14.1/20 | 10.8/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 92→98, token cost down 0%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 446 | human-pages-ai/humanpages (humanpages) humanpages · v1.0.0 | 43.5/100 | F | 0.0/30 | 17.7/30 | 13.7/20 | 12.1/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 92→94, token cost up 5%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 447 | jorgenclaw/nostr-mcp-server (nostr-mcp-server) nostr-mcp-server · v1.0.0 | 43.4/100 | F | 0.0/30 | 16.8/30 | 13.9/20 | 12.7/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 93→91, token cost up 14%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 448 | musepy/genable (genable-mcp) genable-mcp · v1.0.0 | 43.4/100 | F | 0.0/30 | 25.6/30 | 7.1/20 | 10.7/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 100→100, token cost down 0%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 449 | mikusnuz/umami-mcp (@mikusnuz/umami-mcp) mikusnuz-umami-mcp · v1.2.0 | 42.7/100 | F | 0.0/30 | 18.0/30 | 14.2/20 | 10.4/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 100→100, token cost up 15%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 450 | jiezeng2004-design/PatchWarden (patchwarden) patchwarden · v1.7.2 | 42.7/100 | F | 0.0/30 | 17.3/30 | 10.8/20 | 14.6/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 96→96, token cost up 7%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 451 | achiya-automation/safari-mcp (safari-mcp) safari-mcp · v2.15.11 | 42.6/100 | F | 0.0/30 | 18.2/30 | 12.0/20 | 12.4/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 93→89, token cost up 9%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 452 | knowledgepa3/gia-mcp-server (gia-mcp-server) gia-mcp-server · v0.4.4 | 42.5/100 | F | 0.0/30 | 19.5/30 | 9.2/20 | 13.7/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 100→100, token cost up 5%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 453 | Kentucky-ai/opentakeoff (opentakeoff-mcp) opentakeoff-mcp · v0.9.31 | 42.2/100 | F | 0.0/30 | 24.1/30 | 4.0/20 | 14.1/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 100→100, token cost up 2%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 454 | hedging8563/tokenlab-mcp-server (@tokenlabai/mcp-server) tokenlabai-mcp-server · v0.6.17 | 41.9/100 | F | 0.0/30 | 19.8/30 | 10.2/20 | 11.9/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 98→98, token cost up 8%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 455 | agrath/Trello-Desktop-MCP (atlassian-trello-mcp) atlassian-trello-mcp · v1.4.0 | 41.9/100 | F | 0.0/30 | 19.2/30 | 13.3/20 | 9.4/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 95→98, token cost up 12%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 456 | todah-zg/codemagic-mcp (codemagic-mcp-server) codemagic-mcp-server · v1.0.5 | 41.8/100 | F | 0.0/30 | 18.8/30 | 11.8/20 | 11.2/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 98→100, token cost up 4%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 457 | drzamarian/n8n-mcp-community (n8n-mcp-community) n8n-mcp-community · v0.1.2 | 41.6/100 | F | 0.0/30 | 27.7/30 | 1.9/20 | 12.0/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 96%, destructive-action safety rate 0% (baseline only -- no rewrite pass applied).
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 458 | a-y-ibrahim/after-effects-mcp (after-effects-mcp) after-effects-mcp · v1.0.0 | 41.4/100 | F | 0.0/30 | 18.0/30 | 8.5/20 | 14.9/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 100→98, token cost up 5%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 459 | ariekogan/ateam-mcp (@ateam-ai/mcp) ateam-ai-mcp · v0.3.0 | 41.3/100 | F | 0.0/30 | 20.3/30 | 10.3/20 | 10.8/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 100→100, token cost up 1%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 460 | starecz/karea-mcp (karea-mcp) karea-mcp · v0.1.0 | 41.2/100 | F | 0.0/30 | 19.2/30 | 12.0/20 | 10.0/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 98→98, token cost up 7%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 461 | razz-games/razz-mcp (@razzgames/mcp-server) razzgames-mcp-server · v0.2.0 | 41.2/100 | F | 0.0/30 | 18.0/30 | 13.8/20 | 9.4/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 100→100, token cost up 3%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 462 | yuna0x0/anilist-mcp (anilist-mcp) anilist-mcp · v1.4.0 | 40.7/100 | F | 0.0/30 | 16.9/30 | 11.2/20 | 12.6/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 94→98, token cost up 9%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 463 | chessceo/chessceo-mcp (@chessceo/mcp) chessceo-mcp · v0.1.0 | 40.6/100 | F | 0.0/30 | 20.0/30 | 8.4/20 | 12.2/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 98→96, token cost up 3%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 464 | helbertparanhos/cloudflare-mcp-pro (cloudflare-mcp-pro) cloudflare-mcp-pro · v1.1.0 | 40.3/100 | F | 0.0/30 | 20.7/30 | 11.4/20 | 8.2/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 100→100, token cost up 9%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 465 | argosvix/mcp-server (@argosvix/mcp-server) argosvix-mcp-server · v1.1.0 | 38.2/100 | F | 0.0/30 | 20.2/30 | 6.5/20 | 11.5/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 95→93, token cost up 3%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 466 | ashev87/propstack-mcp (propstack-mcp-server) propstack-mcp-server · v1.1.1 | 37.5/100 | F | 0.0/30 | 18.8/30 | 7.7/20 | 11.0/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 94→94, token cost down 1%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 467 | contextium-io/contextium-mcp-server (@contextium/mcp-server) contextium-mcp-server · v0.1.0 | 37.5/100 | F | 0.0/30 | 17.3/30 | 8.7/20 | 11.5/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 95→98, token cost up 11%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 468 | AppVisionOS/apple-search-ads-mcp (apple-search-ads-mcp) apple-search-ads-mcp · v1.0.0 | 37.5/100 | F | 0.0/30 | 18.5/30 | 7.3/20 | 11.7/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 98→100, token cost up 8%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 469 | mikusnuz/meta-ads-mcp (@mikusnuz/meta-ads-mcp) mikusnuz-meta-ads-mcp · v1.3.0 | 37.4/100 | F | 0.0/30 | 18.0/30 | 8.1/20 | 11.4/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 100→98, token cost up 8%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 470 | Notion MCP (official, listTools works unauthenticated) notion-mcp · v1.0.0 | 36.6/100 | F | 0.0/30 | 17.8/30 | 7.8/20 | 11.0/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 91→93, token cost up 2%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 471 | sevalla-hosting/mcp (sevalla-mcp) sevalla-mcp · v1.0.0 | 35.9/100 | F | 0.0/30 | 19.4/30 | 7.3/20 | 9.3/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 100→100, token cost up 3%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 472 | cnghockey/sats4ai-mcp-server (sats4ai-mcp) sats4ai-mcp · v1.0.0 | 35.9/100 | F | 0.0/30 | 21.2/30 | 5.4/20 | 9.3/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 98→96, token cost down 2%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
| 473 | Pantani/tdmcp (@dpantani/tdmcp) dpantani-tdmcp · v0.13.1 | 25.9/100 | F | 0.0/30 | 17.1/30 | 0.0/20 | 8.8/20 | |
Why this scoreAdd explicit identity and permission preflight toolsSelection accuracy 95→93, token cost up 2%, unconfirmed writes 0%→0%.
Highest-impact fix
View full report
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions. |
||||||||
Searching or filtering looks across every assessment, not just the ones shown.
No assessments match those filters. Try a broader search or reset the filters.
Category values are earned points: Safety and Legibility are scored out of 30; Economics and Discoverability out of 20. This batch includes 4 judge/model configurations. Scores are directional and are not directly comparable across configurations.
Initial assessment
Put your MCP
on the board.
Submit a public server URL or npm package. A full Gradable run executes generated tasks against the server and evaluates the outcomes with independent judge CLIs, so assessments run in a controlled environment rather than in your browser.
We’ll run the assessment and follow up with the score. This is the real pipeline—the same one behind the public leaderboard.
Submission received
You’re in the assessment queue.
Thanks—we’ll run the full assessment and follow up with your score.