01Safety
0.0 / 30
What changed in the harness
Selection accuracy 96→98, token cost up 2%, 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
27.5 / 30
03Economics
18.2 / 20
04Discoverability
12.5 / 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.
5 defects 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 |
|---|---|---|
apply_triage |
no_return_description |
Deterministically attach your triage verdicts to their corresponding findings and return the findings array re-sorted by exposure score, highest first. Never suppresses findings. Pass the same 'findings' array you triaged plus a 'verdicts' array of { fingerprint, exposureScore, priority, rationale }. |
generate_cbom |
no_return_description |
Scan a path and return a CycloneDX 1.6 Cryptographic Bill of Materials (CBOM) of the classical cryptographic assets found, for compliance / supply-chain tooling. Reads the filesystem, so it is gated like scan_path over HTTP. |
list_rules |
no_return_description |
Return the quantakrypto detector catalog: every detector id and what it looks for. |
probe_endpoint |
no_return_description |
Actively probe ONE live TLS/SSH endpoint YOU OWN for post-quantum readiness and return the assessment: whether it supports PQC-hybrid key exchange X25519MLKEM768 and its classical certificate posture. REQUIRES an ownership attestation: set i_own_this=true to confirm you are authorized to test the target. Refuses CIDR ranges / wildcards / lists — one host at a time. Performs only a benign, unauthenticated handshake and never modifies the endpoint. NOTE: this is the ONLY quantakrypto MCP tool that opens a network connection; the server is otherwise offline. Over HTTP it is disabled unless the operator sets QUANTAKRYPTO_MCP_ALLOW_NETWORK=1. |
score_delta |
no_return_description |
Compute and return the readiness-score delta and harvest-now-decrypt-later (HNDL) change between two finding sets (e.g. before and after a migration). Pass 'before' and 'after' as arrays of findings from scan_path --format json. |
Selection evidence
Confusable tool pairs.
2 pairs where similar names or overlapping descriptions may send an agent toward the wrong tool.
| Tool A | Tool B | Confidence | Why they collide |
|---|---|---|---|
remediate_findings |
triage_findings |
medium | Both are described as deterministic 'request bundles' (rubric + fix/verdict schema + per-finding metadata) that accept the identical findings array from scan_path, so a vague task like 'process these scan findings' or 'generate the action bundle for these findings' does not clearly distinguish remediation from triage, and an agent may pick whichever tool it encountered first. |
apply_triage |
triage_findings |
medium | Both are triage/verdict tools that consume findings, and the natural phrasing 'apply triage to these findings' maps almost word-for-word to apply_triage while 'triage these findings' maps to triage_findings; a task like 'apply triage to the findings and rank by exposure' can be answered with either tool, and apply_triage also performs the exposure re-sort the task requests. |
Compare the field