
finalize
by achhina
A list of my personal configs
SKILL.md
name: finalize description: This skill should be used when the user asks to "finalize", "/finalize", "ship it", "ship this", "wrap up", "wrap this up", or "finish and commit". Orchestrates the complete workflow for finishing development work including running tests, code review, comment cleanup, and creating a commit. version: 0.1.0
Finalize Development Work
This skill orchestrates the complete workflow for finishing development work by running tests, performing code review, cleaning up comments, and creating a well-thought-out commit.
Purpose
When development work is complete, finishing properly requires several quality gates: verifying tests pass, reviewing code quality, removing spurious comments, and creating a meaningful commit. This skill automates this multi-step workflow to ensure consistent quality before committing changes.
When to Use
Use this skill when the user indicates work is ready to be finalized and committed:
- "finalize" or "/finalize"
- "ship it" or "ship this"
- "wrap up" or "wrap this up"
- "finish and commit"
Workflow Steps
Execute the following steps in order, stopping only if a critical failure occurs that cannot be automatically resolved.
Step 1: Run Tests
Detect and run the appropriate test command for the project:
Detection priority:
- Check for
package.jsonwith test script → usenpm test - Check for
pytest.ini,pyproject.toml, or.pytest files → usepytest - Check for
go.mod→ usego test ./... - Check for
Cargo.toml→ usecargo test - Check for
Makefilewith test target → usemake test
Execution:
# Run the detected test command
<detected-test-command>
On test failure:
- Do not proceed with remaining steps
- Invoke the
superpowers:systematic-debuggingskill to analyze and fix the failures - After fixes are applied, re-run tests to verify they pass
- If tests still fail after debugging attempt, report to user and halt
- Only proceed to Step 2 when all tests pass
Step 2: Code Review
Invoke the code-review agent to analyze recent changes:
Determine review scope:
# Get list of modified files for review
git diff --name-only
Run code review:
Invoke the /code-review skill (uses tdd-workflows:code-reviewer agent for comprehensive analysis):
Use Skill tool: skill="code-review"
This provides comprehensive review covering:
- Security vulnerabilities (OWASP Top 10, injection attacks, secrets)
- Performance issues (algorithms, database queries, memory management)
- Code quality (DRY, SOLID, maintainability, complexity)
- Error handling and resilience
- Testing and testability
- Style and conventions
On review findings:
- Analyze all code review comments and issues identified
- Apply fixes for all issues found (style violations, potential bugs, best practice deviations)
- Re-run tests after applying fixes to ensure nothing breaks
- If any fixes cannot be automatically applied, report to user and halt
- Only proceed to Step 3 when all review issues are resolved and tests still pass
Step 3: Remove Spurious Comments
Invoke the comments skill to clean up obvious and redundant comments:
Use Skill tool: skill="comments"
The /comments skill launches the comment-remover agent which will:
- Scan uncommitted changes for obvious comments
- Remove redundant and self-evident comments
- Preserve valuable documentation (TODO/FIXME, "why" explanations, docstrings)
- Runs in isolated context to reduce main conversation context usage
Step 4: Create Commit
Invoke the commit skill to create a well-thought-out commit message:
Use Skill tool: skill="commit"
The /commit skill will:
- Analyze all changes (git diff and git status)
- Review recent commit messages for style consistency
- Draft a meaningful commit message focused on the "why"
- Create the commit with staged changes
Error Handling
Test failures:
- Attempt automatic fix using systematic-debugging skill
- Verify fix with test re-run
- Halt if tests still fail after debugging
Code review issues:
- Analyze all findings from the comprehensive code review
- Apply fixes for all identified issues (security, performance, style, logic)
- Re-run full test suite after applying fixes
- Halt if critical issues cannot be auto-fixed
- Ensure all high-priority review feedback is addressed before proceeding
Comment cleanup:
- Non-critical, proceed even if no comments removed
Commit creation:
- If no changes to commit, report and complete successfully
- If pre-commit hooks fail, analyze and fix issues, then retry commit
Success Criteria
The workflow completes successfully when:
- All tests pass
- Code review finds no critical issues (or issues are fixed)
- Spurious comments are removed
- A commit is created with a meaningful message
Notes
- This workflow assumes work is complete and ready to commit
- Each step builds on the previous step's success
- Test failures trigger debugging, not just reporting
- Code quality gates prevent committing problematic code
- The final commit represents polished, reviewed, tested work
Score
Total Score
Based on repository quality metrics
SKILL.mdファイルが含まれている
ライセンスが設定されている
100文字以上の説明がある
GitHub Stars 100以上
3ヶ月以内に更新がある
10回以上フォークされている
オープンIssueが50未満
プログラミング言語が設定されている
1つ以上のタグが設定されている
Reviews
Reviews coming soon