version-control
Manage Git repositories and collaborative workflows — branching strategies, commit hygiene, conflict resolution, pull requests, hooks, and .gitignore management. Use when the user requests version control or provides relevant inputs for this workflow.
git clone --depth 1 https://github.com/seb1n/awesome-ai-agent-skills /tmp/version-control && cp -r /tmp/version-control/code-and-development/version-control ~/.claude/skills/version-controlSKILL.md
# Version Control with Git
This skill equips an AI agent to perform the full spectrum of Git-based version control tasks, from initializing a repository and crafting atomic commits to orchestrating branching strategies, resolving merge conflicts, and configuring automation with hooks. It covers both everyday workflows and advanced operations needed for professional team collaboration.
## Workflow
1. **Assess Repository State**: Run `git status`, `git log --oneline -10`, and `git branch -a` to understand the current branch, uncommitted changes, recent history, and available remotes. This context determines which operations are safe to perform.
2. **Stage and Commit Changes**: Group related changes into atomic commits. Write commit messages that start with a concise summary line (50 characters or fewer), followed by a blank line and an explanatory body when the change is non-trivial. Follow the project's commit convention (Conventional Commits, Angular style, etc.) if one exists.
3. **Branch Management**: Create feature, bugfix, or release branches from the appropriate base. Use a consistent naming convention such as `feat/short-description`, `fix/issue-123`, or `release/1.2.0`. Delete stale branches after merging to keep the branch list clean.
4. **Synchronize with Remote**: Fetch and pull regularly to stay up to date. Push feature branches with the `-u` flag on first push. Before merging, rebase or merge the base branch into the feature branch to resolve conflicts early in an isolated context.
5. **Resolve Conflicts**: When conflicts arise, inspect the conflicting files, understand both sides of the change, choose the correct resolution (or combine both), mark the file as resolved with `git add`, and complete the merge or rebase. Always run tests after resolution to confirm correctness.
6. **Automate with Hooks and CI**: Set up Git hooks (pre-commit for linting/formatting, commit-msg for message validation, pre-push for test runs) and ensure `.gitignore` covers build artifacts, environment files, and OS metadata. Integrate with CI pipelines to enforce quality gates on every push.
## Supported Technologies
- **Git** (CLI, libgit2)
- **Platforms**: GitHub, GitLab, Bitbucket, Azure DevOps
- **CLI tools**: `gh` (GitHub CLI), `glab` (GitLab CLI)
- **Hook frameworks**: Husky, pre-commit, Lefthook
- **Branching models**: GitFlow, GitHub Flow, trunk-based development
## Usage
Ask the agent to perform any Git operation by describing what you need in plain language. Examples:
- "Create a feature branch for the new auth module and commit my staged changes."
- "Resolve the merge conflicts in `src/config.ts` and complete the merge."
- "Set up a `.gitignore` for a Python project with a virtual environment."
- "Rebase my branch onto main and force-push the cleaned-up history."
The agent will verify the repository state before every destructive operation and confirm with you before running commands like `reset --hard`, `push --force`, or `rebase` that rewrite history.
## Examples
### Example 1 — Resolving a Merge Conflict
**Scenario**: You are on branch `feat/user-profile` and want to merge `main`, but there is a conflict in `src/utils/format.ts`.
```bash
# Attempt the merge
git merge main
# Output: CONFLICT (content): Merge conflict in src/utils/format.ts
# Inspect the conflict markers
git diff --name-only --diff-filter=U
# src/utils/format.ts
```
The conflicting file contains:
```typescript
function formatDate(date: Date): string {
<<<<<<< HEAD
return date.toLocaleDateString("en-US", { year: "numeric", month: "short", day: "numeric" });
=======
return new Intl.DateTimeFormat("en-US", { dateStyle: "medium" }).format(date);
>>>>>>> main
}
```
**Resolution**: The `main` version uses the modern `Intl.DateTimeFormat` API, which is preferred. Accept that version, remove the conflict markers, and complete the merge:
```typescript
function formatDate(date: Date): string {
return new Intl.DateTimeFormat("en-US", { dateStyle: "medium" }).format(date);
}
```
```bash
# Mark resolved and commit
git add src/utils/format.ts
git commit -m "merge main into feat/user-profile, adopt Intl.DateTimeFormat"
```
### Example 2 — Setting Up a Branching Strategy (GitHub Flow)
**Scenario**: A small team wants a lightweight branching model for continuous deployment.
```bash
# Ensure main is the default and protected branch
gh repo edit --default-branch main
# Create a feature branch from main
git checkout main && git pull
git checkout -b feat/dark-mode
# ... develop and commit ...
git add .
git commit -m "feat: add dark-mode toggle to settings page
Reads user preference from localStorage and applies the
'dark' class to <html>. Falls back to OS preference via
prefers-color-scheme media query."
# Push and open a pull request
git push -u origin feat/dark-mode
gh pr create \
--title "feat: add dark-mode toggle" \
--body "## Summary
- Adds a toggle in Settings that persists to localStorage
- Respects OS-level dark mode preference as default
## Test plan
- [ ] Toggle switches theme immediately
- [ ] Preference persists across page reloads
- [ ] Falls back to OS preference for new users"
# After review and CI pass, merge via GitHub
gh pr merge --squash --delete-branch
```
**Key points**: `main` is always deployable, every change goes through a PR with review and CI, and branches are deleted after merge to avoid clutter.
## Best Practices
- **Write atomic commits.** Each commit should represent one logical change that compiles and passes tests on its own. This makes `bisect`, `revert`, and `cherry-pick` reliable.
- **Never commit secrets.** Add `.env`, credentials files, and private keys to `.gitignore` before the first commit. Use tools like `git-secrets` or `trufflehog` to scan for accidental leaks.
- **Rebase feature branches, merge to main.** Rebasing keeps feature branch history linear and easy to review; merging into main preserves the merge commit as a clear integration point.
- **UseDesign reproducible evaluations for AI agents with representative task sets, explicit rubrics, appropriate graders, baselines, regression gates, and failure analysis. Use when defining agent quality, comparing prompts or models, validating a release, measuring tool-use reliability, investigating regressions, or deciding whether an agent is ready for production.
Design privacy-aware observability for AI agents using traces, spans, structured events, metrics, cost attribution, dashboards, alerts, and investigation workflows. Use when instrumenting an agent, debugging intermittent tool or model failures, defining service-level objectives, analyzing latency or spend, auditing agent decisions, or preparing production monitoring.
Design and verify auditable human oversight, approval gates, escalation paths, and safe state transitions for AI agent workflows. Use when deciding which agent actions require review, adding approve/reject or dual-control flows, preventing unauthorized autonomous effects, creating decision records, reducing rubber-stamping, or recovering safely from rejected, expired, or failed actions.
Design, implement, harden, and verify Model Context Protocol (MCP) servers with precise tool contracts, least-privilege authorization, safe transports, structured errors, and interoperability tests. Use when creating a new MCP server, exposing an API or data source through MCP, reviewing an MCP server design, adding or revising MCP tools, or preparing an MCP server for production.
Design and operate bounded multi-agent workflows with task decomposition, dependency graphs, ownership, handoff contracts, shared-state controls, approvals, recovery, and synthesis. Use when a task contains genuinely independent workstreams, specialized roles, parallel research or implementation, reviewer-worker loops, or coordination problems that one agent should not execute sequentially.
Design and validate model-facing tool definitions with clear names, action-oriented descriptions, bounded JSON Schema parameters, explicit side effects, safe defaults, idempotency, errors, and realistic tests. Use when creating function-calling tools, MCP tools, agent actions, structured tool inputs, or when a model selects the wrong tool, invents arguments, or causes unsafe side effects.
Plan, execute, document, and retest authorized security assessments of AI agents and multi-agent workflows using safe adversarial cases, synthetic identities, canaries, and evidence-based findings. Use when defining red-team rules of engagement, assessing prompt injection or excessive agency, testing tool and identity boundaries, evaluating memory or cross-agent attacks, scoring a campaign, or verifying remediation in an approved environment.
Threat-model and harden AI agents, RAG systems, assistants, and tool-using workflows against direct, indirect, stored, cross-agent, and multimodal prompt injection. Use when reviewing an agent architecture, isolating untrusted content, constraining tools and egress, protecting secrets, adding injection-focused tests, investigating a suspected injection incident, or documenting residual prompt-injection risk.