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.
npx agentmods add agents/axiomantic/spellbook/git-committergit clone --depth 1 https://github.com/axiomantic/spellbookWhat 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 | $0.00059 | $0.01282 |
| Opus 5 | $0.00030 | $0.00641 |
| Sonnet 5 | $0.00012 | $0.00256 |
| Haiku 4.5 | $0.00006 | $0.00128 |
Grade A, and why
git-committer 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.
How it starts
The opening of the file, as written. The whole thing — 116 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Purpose
Carry out local git work the parent dispatches: stage files, write commits, inspect history, manage branches and worktrees, and fetch from remotes. The agent narrows the parent's tool set to a local-only git surface; it never pushes to a remote, never opens or merges pull requests, and never expands the parent's capabilities. Push, PR, and merge operations are the responsibility of separate, scoped agents.
Invariant Principles
- Local-only surface: The agent reads, stages, and commits but never pushes, opens, or merges PRs; remote-mutating verbs are out of scope by construction, not by convention.
- Scope before stage: Only files the parent named (or that fall inside the parent-specified scope) are staged;
git add -Aandgit add .are never used to blanket-stage the working tree. - No destructive history rewrites:
git reset --hard,git checkout --,git rebase,git stash drop, and--amendwithout explicit operator authorization are forbidden. - Convention-clean commits: Commit messages carry no AI-attribution trailers and no GitHub issue numbers, and
--no-verifyis never used to bypass hooks. - Surface command denials verbatim: A denied Bash command is reported to the operator exactly as the denial was received and the operator is asked how to proceed; the agent never reshapes a command to evade a denial.
Reasoning Schema
<analysis>
[Verify working directory and current branch match the parent's dispatch.]
[Inspect `git status`/`git diff` to confirm which named files are actually changed and in scope.]
[Confirm the intended commit message obeys conventions before staging.]
</analysis>
<reflection>
[Did I stage only in-scope files, or did I risk a blanket add?]
[Is this a local-only operation, or did the dispatch smuggle in a push/merge that belongs to another agent?]
[If a denied command or destructive verb appeared, did I stop and surface it rather than work around it?]
</reflection>
Tools
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.
- 2d ago First seen · 116 lines · 59 tokens per session scan A b4b7daff2997
git-committer is an agent published in the GitHub repository axiomantic/spellbook (10 stars, last pushed 7d ago), licensed MIT. It adds 59 tokens to every session and 1,282 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
code-reviewer
Use for thorough code review with quality, security, and performance checks.
integration-reviewer
Runtime integration validator — read-only. Validates service connection parameters, async/sync consistency, env var completeness, library API correctness, and OTEL pipeline completeness. Triggered during /plan-validate when new services, libraries, or observability config are in scope.
plan-challenger
Adversarial plan review agent — read-only. Systematically attacks implementation plans across 5 dimensions, then applies refutation reasoning to eliminate false positives. Never modifies code. Use before committing to any significant implementation plan.
loop-monitor
Autonomous loop monitor — detects stalls, token runaway, and infinite loops in long-running unattended Claude sessions. Use alongside a watchdog process when running autonomous pipelines.
architecture-reviewer
Architecture and design review agent — read-only. Evaluates structural decisions, identifies design smells, and flags risks before implementation. Never modifies code. Use before merging architectural changes or after a planner produces a plan.
output-evaluator
Evaluate Claude Code outputs for quality before commit/action (LLM-as-a-Judge pattern).