01Safety
0.0 / 30
What changed in the harness
Selection accuracy 90→90, token cost up 1%, unconfirmed writes 0%→0%.
Category breakdown
Where the score comes from.
Earned points across the four signals Gradable measures. Safety and Legibility are scored out of 30; Economics and Discoverability are scored out of 20.
02Legibility
26.9 / 30
03Economics
15.2 / 20
04Discoverability
18.0 / 20
Highest-impact fix
Estimated gain +30 pointsAdd explicit identity and permission preflight tools
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions.
Description evidence
Defects and rewrites.
1 defect found across the exposed tool descriptions. Suggested rewrites make purpose, inputs, boundaries, and returns easier for an agent to understand.
| Tool | Defect types | Suggested rewrite |
|---|---|---|
submit_review |
no_return_description |
Call this AFTER you finish integrating an API for a real task - successful or not. Submits your rating, written review, and structured integration report to CLIRank, and returns confirmation that the review was recorded along with the API's updated aggregate score. Takes 10 seconds and improves the score the next agent sees. Without this, scores stay frozen on first impressions. Be honest: 3 stars for 'works fine, nothing special' is more useful than reflexive 5s. Required after any integration sourced via discover_apis or recommend. |
Selection evidence
Confusable tool pairs.
5 pairs where similar names or overlapping descriptions may send an agent toward the wrong tool.
| Tool A | Tool B | Confidence | Why they collide |
|---|---|---|---|
get_api_details |
get_api_docs |
high | Both take a single 'slug' param and sound similar; an agent asking 'get info about the stripe API' could call get_api_details (scores/pricing/capabilities) when it actually needs get_api_docs (integration quickstart), or vice versa. |
get_api_details |
get_reviews |
low | Both take a slug and return API information, but their outputs (static details vs. reviews/ratings) are distinct enough that confusion is unlikely except for vague queries like 'tell me about this API'. |
get_api_docs |
get_package_info |
medium | A task like 'check the latest SDK version and setup info for X' could plausibly be routed to either tool since both surface setup/version-adjacent info, though one is API-level and the other package-registry-level. |
discover_apis |
compare_apis |
low | discover_apis finds options from a need description while compare_apis compares known slugs; overlap only if the agent already has candidate names but phrases the request as an open search. |
get_reviews |
submit_review |
low | Names are related (get vs submit) but described action verbs are clear enough that confusion is unlikely except in truncated instructions like 'review the API'. |
Compare the field