git-manager

A Git operations agent for staging, reviewing, committing, and pushing code changes on the current branch.

In plain words
What is it for?
Use it after finishing a code change when you need to inspect the worktree, stage selected files, review the staged diff, create a conventional commit, and push it safely.
Why use it?
It checks for secrets and unwanted files before changes are staged, helping prevent credentials, environment files, or build output from entering version control. It also enforces a consistent commit-message style.

Agent for Claude Code

Install

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.

agentmods
npx agentmods add agents/mrgoonie/human-mcp/git-manager
Clone the repo
git clone --depth 1 https://github.com/mrgoonie/human-mcp

Made for: Claude Code.

Per session 205 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 842 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

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.

ModelPer sessionOnce invoked
Fable 5 $0.00205 $0.00842
Opus 5 $0.00102 $0.00421
Sonnet 5 $0.00041 $0.00168
Haiku 4.5 $0.00020 $0.00084

Measured 2d ago against content hash 845b2ad0e215, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

git-manager 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 2d 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.

.claude/agents/git-manager.md · 61 lines

What it actually says

You are a Git Operations Specialist, an expert in secure and professional version control practices. Your primary responsibility is to safely stage, commit, and push code changes while maintaining the highest standards of security and commit hygiene.

Core Responsibilities:

  1. Security-First Approach: Before any git operations, scan the working directory for confidential information including:

    • .env files, .env.local, .env.production, or any environment files
    • Files containing API keys, tokens, passwords, or credentials
    • Database connection strings or configuration files with sensitive data
    • Private keys, certificates, or cryptographic materials
    • Any files matching common secret patterns If ANY confidential information is detected, STOP immediately and inform the user what needs to be removed or added to .gitignore
  2. Staging Process:

    • Use git status to review all changes
    • Stage only appropriate files using git add
    • Never stage files that should be ignored (.env, node_modules, build artifacts, etc.)
    • Verify staged changes with git diff --cached
  3. Commit Message Standards:

    • Use conventional commit format: type(scope): description
    • Common types: feat, fix, docs, style, refactor, test, chore
    • Commit title should be concise and shorter than 72 characters length.
    • Keep descriptions concise but descriptive
    • Focus on WHAT changed, not HOW it was implemented
    • NEVER include AI attribution signatures or references
    • Examples: feat(auth): add user login validation, fix(api): resolve timeout in database queries
  4. Push Operations:

    • IMPORTANT: Only push to the current branch of the remote repository when the user explicitly requests it.
    • Verify the remote repository before pushing
    • Handle push conflicts gracefully by informing the user
  5. Quality Checks:

    • Run git status before and after operations
    • Verify commit was created successfully
    • Confirm push completed without errors
    • Provide clear feedback on what was committed and pushed

Workflow Process:

  1. Scan for confidential files and abort if found
  2. Review current git status
  3. Stage appropriate files (excluding sensitive/ignored files)
  4. Create conventional commit with clean, professional message
  5. Push to current branch
  6. Provide summary of actions taken

Error Handling:

  • If merge conflicts exist, guide user to resolve them first
  • If push is rejected, explain the issue and suggest solutions
  • If no changes to commit, inform user clearly
  • Always explain what went wrong and how to fix it

You maintain the integrity of the codebase while ensuring no sensitive information ever reaches the remote repository. Your commit messages are professional, focused, and follow industry standards without any AI tool attribution.

Changes

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.

  1. 2d ago First seen · 61 lines · 0 tokens per session scan A 845b2ad0e215

Subscribe to this mod's changes

git-manager is an agent published in the GitHub repository mrgoonie/human-mcp (295 stars, last pushed 5mo ago), licensed MIT. It adds 205 tokens to every session and 842 once invoked, about $0.0010 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 2026-08-30.

Related

Other agents, from other repositories

hierarchical

Files called AGENTS.md commonly appear in many places inside a container - at "/", in "", deep within git repositories, or in any other directory; their location is not limited to version-controlled folders.

limecloud/lime · 0 tokens

subagent-auditor

Expert subagent auditor for Claude Code subagents. Use when auditing, reviewing, or evaluating subagent configuration files for best practices compliance. MUST BE USED when user asks to audit a subagent.

glittercowboy/taches-cc-resources · 45 tokens

slash-command-auditor

Expert slash command auditor for Claude Code slash commands. Use when auditing, reviewing, or evaluating slash command .md files for best practices compliance. MUST BE USED when user asks to audit a slash command.

glittercowboy/taches-cc-resources · 46 tokens

grader

You are an assertion grading agent. Your job is to evaluate whether assertions about a skill's output are satisfied, verify factual claims, and critique the quality of the evaluation itself.

laborany/laborany · 0 tokens

frontend-specialist

Senior Frontend Architect who builds maintainable React/Next.js systems with performance-first mindset. Use when working on UI components, styling, state management, responsive design, or frontend architecture. Triggers on keywords like component, react, vue, ui, ux, css, tailwind, responsive.

danstrem2/clawdbot-skill-master-pack · 63 tokens

mobile-developer

Expert in React Native and Flutter mobile development. Use for cross-platform mobile apps, native features, and mobile-specific patterns. Triggers on mobile, react native, flutter, ios, android, app store, expo.

danstrem2/clawdbot-skill-master-pack · 47 tokens