frontclippy-audit
Run frontclippy against a local HTML file, live URL, snapshot JSON, suite, or manifest and summarize the findings. Use when the user asks to audit a UI, inspect frontclippy output, emit SARIF, or run
SKILL.md.codex/skills/frontclippy-audit/SKILL.md
GitHubfrontclippy-audit
Use this in the frontclippy repo when the user wants a real audit run instead of a theoretical explanation.
Workflow
- Work from the repo root.
- Choose the narrowest command for the artifact the user provided:
- Local HTML file or live URL:
cargo run -p frontclippy-cli -- check <target>
- Behavioral or keyboard-flow audit:
cargo run -p frontclippy-cli -- trace-check <target>
- Stress or matrix audit:
cargo run -p frontclippy-cli -- matrix-check <target>
- Snapshot JSON:
cargo run -p frontclippy-cli -- analyze <snapshot.json>
- Related snapshots:
cargo run -p frontclippy-cli -- suite-analyze <snapshot-a.json> <snapshot-b.json> [...]
- Manifest:
cargo run -p frontclippy-cli -- manifest-check <manifest.toml>
- Add
--format sarifwhen the user wants CI or code-scanning output. - Add
--config,--baseline,--suppressions, or--write-baselinewhen review state matters. - If no target is provided, use
fixtures/frontclippy.manifest.tomlfor a broad demo orfixtures/checkout-chaos.htmlfor a single-surface demo.
Response Pattern
Return:
- the exact command that ran
- the finding count and highest severity
- whether findings were configured, suppressed, or baselined
- the most relevant next step