@codagent-ai/capture-eval-issues
ACapture noteworthy review violations for the eval framework. Use when validator-run finds review failures — judges violations and saves notable ones to evals/inventory.yml.
Install
agr install @codagent-ai/capture-eval-issues --target claudeWrites 2 files into .claude/skills/, pinned to git-29215f97.
- .claude/skills/capture-eval-issues/SKILL.md
- .claude/skills/capture-eval-issues/judge-prompt.md
Document
name: capture-eval-issues description: >- Capture noteworthy review violations for the eval framework. Use when validator-run finds review failures — judges violations and saves notable ones to evals/inventory.yml. disable-model-invocation: false allowed-tools: Bash, Task, Read
/capture-eval-issues
Capture noteworthy review violations into the eval inventory.
Input
This skill receives one or more review JSON file paths as arguments (space-separated).
Example: /capture-eval-issues validator_logs/review_src_claude@1.0.json validator_logs/review_src_gemini@2.0.json
Procedure
- Read
judge-prompt.mdfrom this skill's directory - Spawn a subagent to judge and capture violations:
- Task tool:
Taskwithsubagent_type="general-purpose",model="sonnet",prompt=judge-prompt content +"\n\nReview JSON files: <file paths>\n\nSkill directory: <this skill's directory path>" - The subagent handles everything: reading files, judging, calling the append script
- Task tool:
- Report the subagent's capture summary (the
CAPTURED:line)
Trustgrade A
- passBody integrity
Whether the stored document is plausibly the kind of file the artifact declares, rather than something fetched by mistake.
- passType matchnot applicable to this artifact type
Whether the artifact is really the kind of thing its metadata claims it is.
- passFreshness
How long since the source repository was last pushed to.
- passPrompt injection
Scans the artifact's own text for instructions aimed at your agent rather than at you.
- passLicense
Whether the source repository declares an SPDX license permissive enough to redistribute.
How the grade is calculated
Each check contributes 0 points when it passes, 1 when it warns, and 2 when it fails. The total maps to a letter:
- Aevery check passed
- Bone warning
- Ctwo warnings
- Dprompt injection or body integrity failed, or three warnings
- Fone of those failed, and something else is wrong
These are automated hygiene checks, not a security audit, and not a dependency or vulnerability scan. A grade of A means nothing was flagged — not that the artifact is safe.
Versions
git-29215f97cbbd2026-07-31