self-audit — independently scanned and version-tracked by SaferSkills.
SaferSkills independently audited self-audit (Agent Skill) and scored it 87/100 (green). The audit ran 55 deterministic rules across Security, Supply Chain, Maintenance, Transparency, and Community; it found 0 high-severity and 3 lower-severity findings. The full rule-by-rule trace and per-finding evidence are below. Free, methodology-open.
Findings & checks · 3 flagged
The text {match} tells the agent to skip the normal "ask the user first" gate. Used adversarially it removes the human-in-the-loop check before destructive or sensitive actions, turning a normally-gated agent into a fire-and-forget executor.
Every scanned point with the score it earned and what moved between them.
First recorded scan — no prior version to compare against.
The primary manifest — the file an agent reads to learn what this artifact does.
This skill requires Python 3.8+ and standard library only. No external packages needed.
To install this skill's dependencies:
pip-compile ./requirements.in
pip install -r ./requirements.txtSee ./requirements.txt for the dependency lockfile (currently empty — standard library only).
Run the analyze-plugin skill against the agent-scaffolders itself and the test fixtures. This is a regression smoke test that verifies the analyzer produces consistent, expected results.
python ./scripts/inventory_plugin.py --path . --format json python ./scripts/inventory_plugin.py --path ./tests/gold-standard-plugin --format json
python ./scripts/inventory_plugin.py --path ./tests/flawed-plugin --format jsonSelf-analysis scanner must confirm:
security_flags = [] (zero security findings in the analyzer itself)issues = [] (zero structural violations)Gold-standard fixture scanner must confirm:
security_flags = [] (zero security findings)issues = [] (zero structural violations)warnings = [] (zero missing components)Flawed fixture scanner must confirm:
security_flags count ≥ 4 (network calls + env access; obfuscated credential is LLM-only)issues count ≥ 1 (bash script violation)warnings count ≥ 2 (missing acceptance criteria + references)./README.md for the full expected findings manifestTo run assertions programmatically:
python ./scripts/assert_audit.py --fixture flawed --json-output <path-to-scan-output.json>tests/gold-standard-plugin/ — should score maturity ≥ L2, zero Critical, at least 2 patterns identifiedtests/flawed-plugin/ — LLM must additionally detect: missing README file tree, missing plugin manifest ⚠️ SELF-AUDIT REGRESSION: [dimension] expected [X] got [Y]
✅ SELF-AUDIT PASSED: [N] scanner checks passed, [M] fixtures validated, [K] 6-phase checks passed~30 seconds. Free. No account. Every finding cites a rule and a line of evidence.