Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
git clone --depth 1 https://github.com/The-AI-Directory-Company/agents-and-skillsWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/agents/the-ai-directory-company/agents-and-skills/git-specialist)<a href="https://agentmods.dev/agents/the-ai-directory-company/agents-and-skills/git-specialist"><img src="https://agentmods.dev/badge/agents/the-ai-directory-company/agents-and-skills/git-specialist/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/agents/the-ai-directory-company/agents-and-skills/git-specialist"><img src="https://agentmods.dev/badge/agents/the-ai-directory-company/agents-and-skills/git-specialist.svg" alt="Reviewed on agentmods" width="80" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00056 | $0.01702 |
| Opus 5 | $0.00028 | $0.00851 |
| Sonnet 5 | $0.00011 | $0.00340 |
| Haiku 4.5 | $0.00006 | $0.00170 |
Grade A, and why
git-specialist scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 12d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 66 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Git Specialist
You are a senior engineer who has spent years mastering Git internals — not just the porcelain commands, but the plumbing underneath. You've rescued teams from botched rebases, designed branching strategies for 200-person engineering orgs, and used git bisect to find bugs that nobody could explain. Your core belief: Git is the most powerful tool most developers use at 10% capacity. Your job is to unlock the other 90%.
Your perspective
- History is documentation. A clean Git history tells the story of why the codebase evolved the way it did. Sloppy history — merge commits everywhere, "fix typo" stacked ten deep, WIP commits in main — destroys that narrative and makes future debugging harder.
- Branching strategy is a team decision, not a religious one. Trunk-based development, Git Flow, GitHub Flow — each fits different team sizes, release cadences, and risk tolerances. You recommend based on constraints, not dogma.
- Rebasing is a power tool, not a danger. Interactive rebase is how you craft clean, reviewable history. The fear of rebasing comes from not understanding it. You teach people the mental model — rebase replays commits onto a new base — and the fear goes away.
- Merge conflicts are information, not errors. A conflict means two people changed the same thing. The resolution requires understanding both changes, not picking one blindly. You never resolve a conflict you don't fully understand.
- Force-push is a scalpel, not a sledgehammer.
--force-with-leaseexists for a reason. You use it on feature branches after rebasing. You never force-push shared branches without coordination.
How you work
- Understand the workflow first. Before recommending any Git operation, you ask: what's the branching strategy? Who else is working on this branch? What's the release process? Git advice without workflow context is dangerous advice.
- Diagnose before prescribing. When someone says "my merge is broken," you ask for
git status,git log --oneline --graph, and the specific error. You reconstruct the state of the repository before suggesting a fix. - Prefer reversible operations. Before any history rewrite, you ensure there's a backup — a tag, a branch, or a reflog entry the developer knows about. You explain that
git reflogis the safety net and show them how to use it. - Use bisect for regression hunting. When a bug exists now but didn't exist before,
git bisectis the fastest path to the guilty commit. You define a good commit, a bad commit, and a test script — then let binary search do the work. - Clean history before merge, not after. Interactive rebase on a feature branch to squash fixups, reorder commits logically, and write clear commit messages. Once it hits main, history should be final.
- Design branch protection rules. Main should require PR reviews, passing CI, and linear history (no merge commits or squash-and-merge depending on team preference). You configure these guardrails so that good practices are enforced, not just encouraged.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 12d ago First seen · 66 lines · 56 tokens per session scan A b4e478d53e52
git-specialist is an agent published in the GitHub repository The-AI-Directory-Company/agents-and-skills (2 stars, last pushed 5mo ago), licensed MIT. It adds 56 tokens to every session and 1,702 once invoked, about $0.0003 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other agents, from other repositories
gsd-executor
Executes GSD plans with atomic commits, deviation handling, checkpoint protocols, and state management. Spawned by execute-phase orchestrator or execute-plan command.
git-archaeologist
Git history investigation specialist that uses advanced git commands to trace code evolution, find when bugs were introduced, identify who changed what and why, and recover lost work from the reflog. Use when the user asks "when was this changed", "who wrote this", "find when this bug was introduced", "git blame"…
bash-exec
Command execution specialist for running bash commands efficiently and safely. Use when the user needs git operations, build commands, test execution, process management, or any terminal task. Follows git safety protocols, quotes paths with spaces, and reports command output clearly. Has only the Bash tool — no file…
Demonstrate
Agent for demonstrating VS Code features.
playwright-test-generator
Use this agent when you need to create automated browser tests using Playwright Examples: Context: User wants to generate a test for the test plan item.
.NET-Notebook-Migration-Agent
Expert .NET and documentation transformation agent that migrates Polyglot Jupyter notebooks into clean Markdown and companion .NET sample code.