humansys/raise/rai-bugfix-fix-enterprise

Claude Code and agent skill by humansys/raise

Trust tier 3
  • community
Source check pending

Checking that the SKILL.md file is still in the repository.

Execute fix tasks with TDD and all validation gates. Phase 5 of bugfix pipeline.

Install

npx vskill@latest install humansys/raise/rai-bugfix-fix-enterprise

SKILL.md

Bugfix Fix

Purpose

Execute the planned fix tasks in strict TDD order: RED (failing regression test) → GREEN (minimal fix) → REFACTOR. Each task verified and committed independently.

Mastery Levels (ShuHaRi)

  • Shu: Follow plan tasks exactly, RED-GREEN-REFACTOR, run all 4 gates
  • Ha: Combine related tasks when gates are stable
  • Ri: Adaptive TDD (property tests, mutation testing)

Context

When to use: After /rai-bugfix-plan has decomposed the fix into atomic tasks.

When to skip: Never — even if the fix seems trivial, follow the plan.

Inputs: Bug ID, work/bugs/{issue_key}/plan.md with atomic tasks.

Expected state: On bug branch. Plan artifact exists. All current gates pass.

Steps

Step 1: Execute Tasks in Order

Per task from plan.md:

  1. RED — Write a failing test that defines expected behavior
  2. GREEN — Write minimal code to make the test pass
  3. REFACTOR — Clean up while keeping tests green

Run ALL four gates (test + lint + format + type check) after each task via the CLI — NEVER run the test command directly (keeps context clean):

bash
rai gate check gate-tests --scope {changed_test_path} # fast check, scoped
rai gate check gate-lint
rai gate check gate-format
rai gate check gate-types

Fall back to rai gate check gate-tests (unscoped) only when the scope path is ambiguous.

If verification fails: fix and re-verify (max 3 attempts before escalating).

Step 2: Commit & Present

Per completed task:

  1. Commit with the message from the plan
  2. Present to the human: what was completed, files changed, verification results
  3. Wait for acknowledgment before continuing
Step 3: Iterate or Finalize
ConditionAction
More tasks remainReturn to Step 1
All tasks completeRun rai gate check --point before:bug:close, present summary
Task blockedDocument blocker, escalate to human

<verification> All tasks committed. All four gates pass. Bug no longer reproduces. </verification>

<if-blocked> 3 attempts without fix → document partial state, create follow-up issue. </if-blocked>

Step 4: Quality Review of Implementation

After all tasks complete and all four gates pass, invoke /rai-quality-review on the changed files:

bash
git diff --name-only $(git merge-base HEAD <parent-branch>)..HEAD
VerdictAction
PASS / PASS WITH RECOMMENDATIONSProceed to /rai-bugfix-review
FAILAddress critical findings, re-run gates, then proceed

<verification> QR verdict is PASS or PASS WITH RECOMMENDATIONS. No critical semantic bugs in fix. </verification>

Output

ItemDestination
Code + commitsOn bug branch

STOP HERE. Return your summary to the orchestrator. Do NOT invoke any further skill.

Quality Checklist

  • [ ] TDD cycle followed per Step 1 — RED proves bug, GREEN minimal fix, REFACTOR cleans up while keeping tests green
  • [ ] Regression test not skipped; root cause addressed (not symptoms)
  • [ ] All four gates pass after each task (test, lint, type check, format)
  • [ ] Each task committed independently
  • [ ] Human acknowledged each task before proceeding
  • [ ] Bug no longer reproduces after fix
  • [ ] QR verdict is PASS or PASS WITH RECOMMENDATIONS before handoff to review

References

  • Gate: gates/gate-code.md
  • Previous: /rai-bugfix-plan
  • Next: /rai-quality-review → /rai-bugfix-review

Backlog tracker

Engine behavior (RAISE-16987): The engine checks ExecutionContext to determine mode:

  • mode: pipeline — engine active; apply_phase_transition moves the tracker. No skill call required.
  • mode: standalone — no active run; tracker is not moved automatically.

If standalone (mode == standalone via ExecutionContext):

bash
rai backlog transition {key} --point after:bugfix:fix

Security

  • Security scanPassed, 95 of 100Pattern checks on the skill's files.
  • Model reviewPassed

Quality evaluation

  • Task evaluationNot runNo evaluation yet. Run one in Skill Studio.

Source

Popularity

Repository stars
71
Repository forks
8
7-day trend score
21

Stars and forks are for the whole GitHub repository, not this skill alone. Not refreshed yet.

Works with

Any agent vskill installs to: 56 targets in all. These 39 have a page on this site.

  • Claude Code
  • Cursor
  • GitHub Copilot
  • Windsurf
  • Codex
  • Gemini CLI
  • Cline
  • Amp
  • Roo Code
  • Goose
  • Continue
  • Kilo Code
  • Junie
  • Trae
  • OpenHands
  • Qwen Code
  • AI Agent
  • Tabnine Chat
  • Sourcegraph Cody
  • Amazon Q
  • Firebase Genkit
  • Sweep
  • Devin
  • Phind
  • Mentat
  • Claude Desktop
  • Aider
  • GPT Pilot
  • Plandex
  • bolt.new
  • v0
  • Replit Agent
  • Anse
  • TypingMind
  • BetterChatGPT
  • LibreChat
  • Warp AI
  • Pieces
  • Fig AI
Embed badge
vskill VERIFIED badge
[![vskill VERIFIED](https://verified-skill.com/api/v1/skills/humansys/raise/rai-bugfix-fix-enterprise/badge)](https://verified-skill.com/skills/humansys/raise/rai-bugfix-fix-enterprise)

Something wrong with this skill or its page? Tell us.