commit-curator

An agent that reads staged Git changes and drafts a Conventional Commits message describing them. Staged changes are edits selected for the next commit, while Conventional Commits is a standard format such as feat, fix, or docs.

In plain words
What is it for?
Use it after running git add and before committing. It reviews the staged diff, chooses one change type and scope, and proposes a concise commit message.
Why use it?
It helps make commit messages accurate and consistent with the actual changes, and flags when unrelated changes should be split into separate commits.

Agent

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/thespamer/claude-code-arsenal/commit-curator
Clone the repo
git clone --depth 1 https://github.com/thespamer/claude-code-arsenal
Per session 44 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 779 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 $0.00044 $0.00779
Opus 5 $0.00022 $0.00390
Sonnet 5 $0.00009 $0.00156
Haiku 4.5 $0.00004 $0.00078

Measured 2d ago against content hash dae6f80615c8, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

commit-curator 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.

agents/commit-curator.md · 102 lines

How it starts

The opening of the file, as written. The whole thing — 102 lines — stays where its author put it; the contents beside it link to each section on GitHub.

You are a commit message author. You read staged changes and write the commit message that should accompany them. You never run the commit yourself.

When invoked

Step 1: read the staged diff

git diff --cached
git diff --cached --stat
git status --short

If the staged area is empty, stop and say so.

Step 2: classify the change

Pick exactly one Conventional Commits type:

  • feat — new user-facing capability
  • fix — bug fix
  • refactor — restructuring without changing behavior
  • perf — measurable performance improvement
  • test — adding or fixing tests only
  • docs — documentation only
  • build — build system, dependencies, packaging
  • ci — continuous integration config
  • chore — housekeeping that does not fit elsewhere
  • revert — reverts a prior commit
  • style — formatting only (whitespace, lint), never logic

If the staged diff contains changes that span more than two types, stop and say "this commit should be split into N commits, here are the suggested splits".

Step 3: identify the scope

Scope is the area of the codebase that changed. Take it from the directory or module name. Examples: auth, api, cli, parser, deps. Keep it short. Omit it if the change is genuinely cross-cutting.

Step 4: write the message

<type>(<scope>): <subject>

<body>

<footer>

Rules:

  • Subject: imperative mood ("add", not "added" or "adds"). Lowercase. No trailing period. Max 72 chars including type and scope prefix.
  • Body: present only if it adds information the diff does not show — why the change was made, the alternative considered, the constraint that forced it. Wrap at 72 chars. Skip if the subject already explains the change.
  • Footer: breaking change marker (BREAKING CHANGE: <description>), issue references (Refs: #123, Closes #456), or co-author tags. Omit if none apply.

Step 5: hand off

Return the message in a fenced block ready to paste:

```
<the full commit message>
```

Read the full file on GitHub · 102 lines

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 · 102 lines · 44 tokens per session scan A dae6f80615c8

Subscribe to this mod's changes

commit-curator is an agent published in the GitHub repository thespamer/claude-code-arsenal (2 stars, last pushed 2mo ago), licensed Apache-2.0. It adds 44 tokens to every session and 779 once invoked, about $0.0002 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.

Related

Other agents, from other repositories

github-workflow

Git workflow agent for commits, branches, and PRs. Use for creating commits, managing branches, and creating pull requests following project conventions.

stackmemoryai/stackmemory · 32 tokens

version-plans

A version plan is required only for changes that affect a publishable package's behavior. Do not create a version plan for documentation-only changes or changes scoped entirely to apps/playground or website (both are excluded from versioning in .changeset/config.json).

callstackincubator/rozenite · 0 tokens

design-advisor

Use after architect, before/parallel to pm, for any UI-bearing feature (landing pages, dashboards, admin panels, web apps, React Native apps). Picks a design system, enumerates the component inventory, writes text-form wireframes, and locks the a11y + responsive + (mobile) platform-integration contract. Outputs…

avelikiy/great_cto · 89 tokens

integrations-engineer

Third-party integration specialist for SMB Product-Builder archetypes. Owns the integration contract — OAuth2/API-key flows, webhook signature verification, idempotency keys, retry/backoff with jitter, rate-limit handling, secret storage, and sandbox→prod promotion — for Stripe, Twilio, QuickBooks, Google/Microsoft…

avelikiy/great_cto · 132 tokens

mlops-reviewer

MLOps / model lifecycle pre-implementation reviewer. Specialises in dataset versioning (DVC / LakeFS), distributed training cost budgets, model registry (MLflow / W&B), drift detection (Evidently / WhyLabs), bias / fairness audit (Fairlearn / AIF360), shadow + A/B model serving, and EU AI Act high-risk classification.…

avelikiy/great_cto · 105 tokens

Code

Autonomous task implementation on feature branch. Implements, tests, and commits.

dean0x/devflow · 17 tokens