Docs/findings quality evidence roadmap#263
Merged
Merged
Conversation
Raise fact_coverage to 80%, add fleet absolute-risk gates and auxiliary v2 gate parity, extend MCP/CLI trust surfaces, and apply ruff format to files that were failing CI on PR #565.
The action defaults ci-trust to true, which overrides fail-on-critical false and applies min-score 70 — intentional for real CI but breaks the validate workflow that only checks report artifacts from the demo server.
When the semantic model is unavailable, the analyzer emits a coverage skip finding that should not count as a credential detection in MCTS-T-1022 fixtures.
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
The Validate Action failure was a mismatch between the new action default and the smoke test target.
What broke: action/action.yml now defaults ci-trust to true, which adds --ci-trust (enforce mode, min-score 70, fail-on-critical). The validate workflow scans examples/vulnerable-mcp-server/server.py and only set fail-on-critical: "false", but --ci-trust overrides that — so the scan correctly reported score 2/100 and 3 critical findings, exited 1, and never copied SARIF/HTML to the workspace (hence the artifact upload warnings).
Fix: Explicitly opt out in .github/workflows/action-validate.yml:
ci-trust: "false"
fail-on-critical: "false"