원클릭으로
finishing-branch
Use when completing development on a feature branch -- verifies quality, presents completion options, and cleans up
Codex 또는 Claude로 설치 이 Prompt를 복사해 Codex, Claude 또는 다른 어시스턴트에 붙여 넣으면 Skill 페이지를 검토하고 설치를 진행할 수 있습니다.
메뉴
Use when completing development on a feature branch -- verifies quality, presents completion options, and cleans up
Codex 또는 Claude로 설치 이 Prompt를 복사해 Codex, Claude 또는 다른 어시스턴트에 붙여 넣으면 Skill 페이지를 검토하고 설치를 진행할 수 있습니다.
SOC 직업 분류 기준
Guide for structuring, parsing, and validating XML-like tags in LLM inputs, system prompts, outputs, and agent-to-agent communication. Use this skill when designing prompt templates, parsing structured content from text, or handling nested tag boundaries.
Implementation reference library for compound engineering workflows. For workflow orchestration, use slash commands under `commands/workflows/` (constitution, plan, work, review, compound). This skill provides stack-specific fix recipes those workflows consume.
Browser automation using Vercel's agent-browser CLI. Use when you need to interact with web pages, fill forms, take screenshots, or scrape data. Alternative to Playwright MCP - uses Bash commands with ref-based element selection. Triggers on "browse website", "fill form", "click button", "take screenshot", "scrape page", "web automation".
Build applications where agents are first-class citizens. Use this skill when designing autonomous agents, creating MCP tools, implementing self-modifying systems, or defining architecture artifacts around deletion tests, interfaces, seams, adapters, and explicit contracts.
Run a comprehensive scored audit of agent-native architecture principles
This skill should be used before implementing features, building components, or making changes. It guides exploring user intent, approaches, and design decisions before planning. Triggers on "let's brainstorm", "help me think through", "what should we build", "explore approaches", ambiguous feature requests, or when the user's request has multiple valid interpretations that need clarification.
| name | finishing-branch |
| description | Use when completing development on a feature branch -- verifies quality, presents completion options, and cleans up |
| model | claude-sonnet-4.6 |
| platforms | {"copilot":{"model":"gpt-5.3-codex"},"opencode":{"model":"openrouter/moonshotai/kimi-k2.6"}} |
A structured skill for completing work on a feature branch. Ensures all quality checks pass, presents clear options for what to do next, and handles cleanup.
Announce at start: "I'm using the finishing-branch skill to complete this work."
workflows:work Phase 4/lfg pipelineRun all quality checks and report results:
# 1. Run full test suite (detect project test command)
# npm test | pytest | cargo test | phpunit | go test | etc.
# 2. Run linting (detect project linter)
# eslint | ruff | clippy | pint | etc.
# 3. Check for uncommitted changes
git status
# 4. Check for untracked files that should be committed
git ls-files --others --exclude-standard
Report results clearly:
Do NOT proceed if tests or linting fail. Fix issues first, then return to Step 1.
If there are uncommitted changes:
git add <relevant files>
git commit -m "feat(scope): final changes before merge
[description of what's included]"
Stage files selectively -- do not blindly git add . if there are files that should not be committed.
Present the user with clear options:
Option A: Create a Pull Request (recommended for team projects)
Option B: Merge to default branch (for solo projects or pre-approved changes)
Option C: Keep the branch (for work that needs more review)
Option D: Discard the branch (if work is no longer needed)
Ask the user which option they prefer. Do not assume.
When invoked by an orchestrator without user interaction available: Default to Option A (Create PR). Do not stall waiting for user input -- proceed with creating the PR automatically.
git push -u origin [branch-name]
Generate a PR/MR description:
## Summary
- What was built and why
- Key decisions made
## Changes
- [List major changes with file references]
## Testing
- Tests added/modified
- Manual testing performed
## Post-Deploy Monitoring & Validation
- **What to monitor:** [logs, metrics, dashboards]
- **Expected healthy behavior:** [what success looks like]
- **Failure signals:** [what to watch for]
- **Validation window:** [how long to monitor]
## Screenshots
[If UI changes, include before/after screenshots]
---
[](https://github.com/aegntic/compound-engineering)
# Update default branch
git checkout [default-branch]
git pull origin [default-branch]
# Merge feature branch
git merge [feature-branch] --no-ff -m "Merge [feature-branch]: [description]"
# Push
git push origin [default-branch]
# Clean up local branch
git branch -d [feature-branch]
git push -u origin [branch-name]
echo "Branch pushed. Ready for review."
# Confirm with user first!
git checkout [default-branch]
git branch -D [feature-branch]
echo "Branch discarded."
If a git worktree was used:
# Return to main working directory
cd [main-worktree-path]
# Remove the worktree
git worktree remove [worktree-path]
# Prune worktree references
git worktree prune
Update plan status (if a plan file was used):
status: active to status: completed in the plan's YAML frontmatter/workflows:work sessionsUpdate execution session (if STATE.md exists):
status: completed in STATE.mdPresent a clear completion summary:
## Branch Complete: [branch-name]
- Action taken: [PR created / Merged / Kept / Discarded]
- Tests: [X passed]
- Commits: [N commits]
- Files changed: [N files]
- PR/MR: [link if created]
- Worktree: [cleaned up / N/A]
- Plan status: [updated to completed / N/A]