debug
Investigation-first debugging — gather evidence, form confirmed root-cause hypothesis, hand off to fix mode with diagnosis file.…
Verify and resolve review findings, account for every finding in a terminal bucket, and prove closure.
$ npx -y skills add Borda/AI-Rig --skill rig-remediate --agent claude-codeHow it fires
How this skill gets triggered: by you, by Claude, or both.
/rig-remediateContext preview
The summary Claude sees to decide when to auto-load this skill.
Verify and resolve review findings, account for every finding in a terminal bucket, and prove closure.
name: Rig Remediate description: Verify and resolve review findings, account for every finding in a terminal bucket, and prove closure.
> Read [Evidence rules](../../references/evidence.md) and [Commit handoff](../../references/commit-handoff.md).
1. **Verify before acting.** Treat findings as untrusted until checked against current source. A finding whose line, symbol, or invariant has changed is `stale`; do not "fix" a stale claim. 2. **Start the run.** `rig_evidence` with `action: "start"`, `skill: "rig-remediate"`, and the affected paths. 3. **Bucket every finding.**
4. **Assign owners.** Use `rig-implementer` for changes, `rig-verifier` for adversarial verification. One writer per path. 5. **Fix and verify.** Implement the smallest correct change. Add or update the test that proves the finding. Re-run the focused check per finding. 6. **Reconcile and gate.** Refresh the run scope with `rig_evidence` (`action: "refresh"`) after the last edit, then run the configured gates with `rig_gate` (`action: "run"`). If a gate rewrites files, refresh again and rerun it. 7. **Re-review fixed areas.** An independent reviewer checks the changed source against the original finding (include correctness). Record that review with `rig_evidence` (`action: "record"`, `reviewer`) after the refresh, so the evidence binds to the current digest. The implementer's own report is not verification. For the re-review, require axes including `correctness`. 8. **Reconcile.** Every finding ID must appear in exactly one terminal bucket. Write the final statuses with `rig_findings`; counts must match the input inventory. 9. **Finalize.** `rig_evidence` with `action: "finalize"`; an open finding, a stale review digest, or a failed gate blocks completion.
Report closure counts, per-finding status, changed paths, gate receipts, remaining risks, and the commit handoff.
Practical agent workflows for Python, ML, and open-source maintenance. AI-Rig turns recurring work—scoping a change, reproducing a bug, reviewing a pull request, running an experiment, or checking release readiness—into explicit workflows with specialist
Investigation-first debugging — gather evidence, form confirmed root-cause hypothesis, hand off to fix mode with diagnosis file.…
TDD-first feature development — crystallise API as a demo test, drive implementation to pass it, run quality stack and…
Reproduce-first bug resolution — capture bug in failing regression test, apply minimal fix, run quality stack and review loop.…
Analysis-only planning — classify and scope a task without writing code; outputs a structured plan to .plans/active/. TRIGGER…
Test-first refactoring — audit coverage, add characterization tests, apply changes with safety net, run quality stack and review…
Multi-agent code review of local Python files, directories, or the current git diff covering architecture, tests, performance,…