SKILL.md
Brooks-Lint — Full Sweep & Auto-Fix
When to Use
Use this skill when you need full-sweep mode: runs a unified analysis across all quality dimensions — code decay, architecture, tech debt, and test quality — then applies fixes directly to the codebase. Safe changes are auto-applied; risky changes are confirmed before execution. Drawing on twelve classic...
Setup
- Read
../_shared/common.mdfor the Iron Law, Project Config, Report Template, and Health Score rules - Read
../_shared/source-coverage.mdfor book-level coverage, exceptions, and tradeoffs - Read
../_shared/decay-risks.mdfor production risk symptom definitions - Read
../_shared/test-decay-risks.mdfor test risk symptom definitions - Read
sweep-guide.mdin this directory for the unified scan and fix process
Process
If the user has not specified a project or directory: apply Auto Scope Detection from ../_shared/common.md to determine the review scope before proceeding.
- Show pre-flight consent notice and wait for the user's one-time approval (Step 0 of the guide)
- Enumerate scope and initialize the
unresolvable/noncriticalrounds/fix_logstate (Step 1 of the guide) - Run the four dimensions in sequence — review, test, debt, audit — each scanning, classifying, applying Safe + Extended-Safe fixes, and verifying via the project test command (Steps 2–5 of the guide)
- Iterate: re-scan modified files + same-module + static consumers; converge on a clean round, retire 3-retry failures to the
unresolvableset, cap non-critical rounds at 3 (Step 6 of the guide) - Aggregate residual and unresolvable items and output the Full Sweep Report (Steps 7–8 of the guide)
Mode line in report: Full Sweep
Limitations
- Use this skill only when the task clearly matches its upstream source and local project context.
- Verify commands, generated code, dependencies, credentials, and external service behavior before applying changes.
- Do not treat examples as a substitute for environment-specific tests, security review, or user approval for destructive or costly actions.