Skip to content
cccskills
FreePublished on GitHub

git-workflow-and-versioning

Structure git work with atomic commits, conventional messages, short-lived branches, and pre-commit hygiene. Load when committing, branching, resolving merge conflicts, organizing parallel work, or reviewing git history. Also triggers on "git workflow", "commit message", "conventional commits", "atomic commit", "branch strategy", "git worktree", "how should I commit this". Does not replace project-specific hook policies — complements them.

The English summary is being prepared. Showing the original description.

See how to install

Files included (2)

  • SKILL.md5.1 KB
  • references/examples.md1.7 KB

SKILL.md (original)

Review the instructions the agent will receive before you install.

Git Workflow and Versioning

You keep git history reviewable, reversible, and honest. Commits are save points; messages document intent.

Hard Rules

One logical change per commit — never mix feature + refactor + formatting. Commit messages explain why, not only what the diff shows. Never commit secrets — scan staged diff for password, secret, api_key, token patterns. Never force-push to shared main/master without explicit user request. Tests should pass before commit when the project has a test command.


Workflow

Step 1 — Review what's changing

Run git status and git diff (staged if committing). Confirm scope matches one logical change.

Step 2 — Pre-commit hygiene

Run project tests/lint if applicable. Reject staged secrets. Split if mixed concerns.

Step 3 — Write the message

<type>: <short imperative description>

<optional body — why, tradeoffs, links to spec/task>

Types: feat | fix | refactor | test | docs | chore

Step 4 — Commit or advise the user

If the user's environment allows commits, commit. Otherwise output the exact message for them to run.

Step 5 — Summarize for reviewers

Emit a short change summary (see Output Format).

Step 6 — Finish the branch (when the task is complete)

Verify tests are green before offering to finish. Present exactly these options: 1) merge locally, 2) push and open a PR, 3) keep the branch as-is, 4) discard. Destructive cleanup (branch delete, worktree removal) requires the user to explicitly confirm the discard choice — never delete on an inferred or unconfirmed "yes."


Branching defaults

  • Trunk-based: keep default branch deployable; merge feature branches within 1–3 days.
  • Naming: feature/<slug>, fix/<slug>, chore/<slug>, refactor/<slug>
  • Parallel agents: prefer git worktree add over constant branch switching.
  • Incomplete features: feature flags on main beat month-long branches.

Gotchas

  • Squashing everything at the end destroys the slice narrative — commit incrementally.
  • Formatting-only commits mixed with behavior changes make review and bisect painful.
  • Missing .gitignore until .env lands in history.
  • git reset --hard on shared branches loses others' work — warn first.
  • Never delete a branch or worktree as part of "finishing up" without the user explicitly picking that option — an assumed cleanup can discard unmerged work.

Common Rationalizations

ExcuseReality
"I'll commit when the feature is done"One giant commit can't be reviewed or reverted safely.
"Message doesn't matter"History is documentation for future you and agents.
"I'll squash later"Prefer clean incremental commits from the start.
"Branches are overhead"Short-lived branches isolate work; long-lived branches are the cost.
"Tests pass, I'll just clean up the branch"Cleanup (merge/discard) is the user's call — present options, don't act unilaterally.

Output Format

## Git summary

Commits recommended:
- `<type>: <description>`

CHANGES:
- path: one-line what/why

NOT TOUCHED (intentional):
- path: reason out of scope

CONCERNS:
- [risk or question for human]

Examples

<examples> <example> <input>I'm done with the validation endpoint. Ready to commit.</input> <output> Staged review: one logical change (POST /validate + tests). Message: ``` feat: add skill validation endpoint

Exposes validate-skills checks for CI; refs task validate-api.

Run: `npm test` → pass, then commit.
    </output>
  </example>
</examples>

---

## Verification

- [ ] Commit message follows convention
- [ ] Scope is single logical change
- [ ] Handoff completed if user triggered via commit/push
- [ ] No secrets in committed files

## Red Flags

- All work squashed into one commit losing slice narrative
- Formatting-only changes mixed with behavior commits
- .env or secrets committed before gitignore exists
- Conventional commit type mismatches actual change nature

## Prune Log
Last pruned: 2026-07-09
- Added branch-finishing flow (merge/PR/keep/discard) with confirm-before-destroy gate (agent-loom Phase 4, obra/superpowers)


## Impact Report

Git workflow: [commit advised / committed / review-only] Commits: N | Concerns flagged: N

Reviews

No reviews yet. Tell others what you thought.

Comments

No login needed. Share how it worked for you, ask a question, or note where it did not run.

No comments yet. Be the first to write one.

Comments are published right away. Do not include personal information such as email addresses or phone numbers. Up to 2 URLs.

0 / 1,000 characters

This form is protected by reCAPTCHA, and the Google Privacy Policy and Terms of Service apply.

More skills from this repository

Put on the adversarial hat and systematically attack any document, plan, strategy, or idea to expose its weakest points before commitment. Structured devil's advocate with red team rigour — not pessimism, but evidence-based critique across three phases: diagnostic (are claims accurate?), creative (is the problem artificially constrained?), challenge (are solutions robust?). Load when the user asks to stress test a document, red team this plan, poke holes in this, devil's advocate this, challenge my assumptions, or when product-soul, brainstorming, prd-writing, or inversion calls for adversarial review. Also triggers on "what am I missing", "what could kill this", "find the flaws", or "critique this rigorously".

The English summary is being prepared. Showing the original description.

dvy1987/agent-loom3Updated Aug 8, 2026

Design execution structure for decomposed processes: single agent or multi-agent topology. Load when user says "design an agent for this", "what agent structure do I need", "architect this", "should this be multi-agent", "what's the right execution structure", "agent topology", "how should agents be organized". Takes process-decomposer output as primary input. If triggered directly without a process entry, calls process-decomposer first.

The English summary is being prepared. Showing the original description.

dvy1987/agent-loom3Updated Aug 8, 2026

Internal skill. Called by setup-evaluation after a PASS. Launches agents from a validated architecture spec using Claude Code / Ampcode native parallelism (Task tool). Does NOT generate scripts or SDK code — it outputs structured spawn instructions that the platform executes natively. Never invoked directly by the user. Never launches without a setup-evaluation PASS.

The English summary is being prepared. Showing the original description.

dvy1987/agent-loom3Updated Aug 8, 2026

Sync library skills from an agent-loom upstream repo into this project's .agents/skills while preserving project-local and forked skills. Load when the user asks to sync agent-loom, update skills from upstream, rsync from ../agent-loom, pull new library skills, upgrade installed skills, or refresh the .agents folder without losing custom project skills. Also triggers on "sync skills from agent-loom", "update my agent skills", "pull skill library updates", or "merge agent-loom improvements into this repo".

The English summary is being prepared. Showing the original description.

dvy1987/agent-loom3Updated Aug 8, 2026

Instrument a shipped product's AI agents with tracing and observability so you can see what they did, why outputs happened, and what each run cost. Plain-language primer plus free-tier-first backend selection (Langfuse, Phoenix, LangSmith, Braintrust) and OpenTelemetry/OpenInference instrumentation. Load when the user asks to add observability, add tracing, instrument my agents, see what my agent is doing in production, set up Langfuse or Phoenix or LangSmith, debug why my agent gave a bad answer, or track LLM cost per request. Also fires when agent-system-architecture or setup-evaluation requires an observability plan for an agent-chain product. NOT for tracing the coding agent itself — that is run-trace. Precondition for runtime-learning-loop.

The English summary is being prepared. Showing the original description.

dvy1987/agent-loom3Updated Aug 8, 2026

Run a structured retrospective after development-phase runs of your product's agents — interview the owner in plain language about what went well and poorly, draft ranked improvement hypotheses, then design and run small n=1/n=2 experiments with pre-declared success criteria, guardrails, stop conditions, and a cost/ROI kill-switch. Load when the user says how did that run go, retro this run, the agent output was bad, what should we improve, draft hypotheses, run a small experiment, or after repeated dev runs of an agentic system produce uneven quality. Priority: output quality over performance over cost, each with diminishing-returns stops. NOT a product A/B test (experimentation), NOT coding-agent harness repair (harness-evolution), NOT production-scale learning (runtime-learning-loop).

The English summary is being prepared. Showing the original description.

dvy1987/agent-loom3Updated Aug 8, 2026

See all skills by dvy1987

Report a problem with this skill