Agent skill

tools-bug-scan

Run the repository-native bug scanner (`pnpm bug-scan` / `pnpm bug-scan:changed`), summarize findings by severity and rule, and propose concrete next fixes or targeted rescans.

Stars 163
Forks 31

Install this agent skill to your Project

npx add-skill https://github.com/majiayu000/claude-skill-registry/tree/main/skills/other/other/tools-bug-scan

SKILL.md

Bug Scan Tool

Use this skill when the user wants a fast, deterministic bug-pattern sweep using the in-repo scanner.

Commands

  • Full/targeted scan:
    • pnpm bug-scan [path ...]
  • Changed-files scan:
    • pnpm bug-scan:changed
  • Machine-readable output:
    • pnpm bug-scan -- --format=json [path ...]
    • pnpm bug-scan -- --changed --format=json
    • pnpm bug-scan -- --changed --format=json --fail-on=none --business-scope=<BIZ> --idea-artifact=docs/plans/<slug>/bug-scan-findings.user.json
  • Rule discovery and tuning:
    • pnpm bug-scan:rules
    • pnpm bug-scan -- --changed --fail-on=critical
    • pnpm bug-scan -- --only-rules=<ruleA,ruleB>
    • pnpm bug-scan -- --skip-rules=<ruleA,ruleB>
    • pnpm bug-scan -- --write-baseline=tools/bug-scan-baseline.json --fail-on=none
    • pnpm bug-scan -- --baseline=tools/bug-scan-baseline.json

Default Workflow

  1. Choose scope:
    • If user asks for changed work, run pnpm bug-scan:changed.
    • If user provides files/directories, pass them to pnpm bug-scan.
    • If unspecified, prefer changed-files first to keep signal high.
  2. Run scanner once.
  3. If findings exist, summarize by:
    • severity counts (critical, warning)
    • rule IDs
    • top file locations (path:line:column)
  4. Recommend a concrete next action set:
    • immediate fixes for critical
    • optional cleanup queue for warning
    • exact rerun command after fixes
  5. If the scan is part of post-build reflection, write the idea artifact into the active plan directory so startup-loop:generate-process-improvements can ingest findings as idea candidates.

Output Contract

Always report:

  • Scan scope used.
  • Finding totals by severity.
  • Top findings with file references.
  • Exit status implication:
    • 0 = no findings
    • 1 = findings present

Guardrails

  • Do not install external scanners or mutate hooks/shell profiles in this skill.
  • Do not fail the task if scanner reports findings; report and propose fix order.
  • If scanner command fails for runtime/tooling reasons, report the error and suggest the minimal unblocking command.

Escalation

Escalate to planning (lp-do-fact-find / lp-do-build) when:

  • the same rule repeatedly appears across many files,
  • findings indicate systemic pattern debt,
  • user asks to convert findings into an implementation plan.

Expand your agent's capabilities with these related and highly-rated skills.

Didn't find tool you were looking for?

Be as detailed as possible for better results