Borrowing it
Nothing to install: this file belongs to tallclub/matimo. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.
curl -O https://raw.githubusercontent.com/tallclub/matimo/main/.claude/skills/split-commits/SKILL.mdgit clone --depth 1 https://github.com/tallclub/matimoWrote 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/skills/tallclub/matimo/split-commits)<a href="https://agentmods.dev/skills/tallclub/matimo/split-commits"><img src="https://agentmods.dev/badge/skills/tallclub/matimo/split-commits/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/skills/tallclub/matimo/split-commits"><img src="https://agentmods.dev/badge/skills/tallclub/matimo/split-commits.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 | $0.00079 | $0.02071 |
| Opus 5.5 | $0.00032 | $0.00828 |
| Sonnet 5.5 | $0.00016 | $0.00414 |
| Haiku 4.5 | $0.00008 | $0.00207 |
Grade A, and why
split-commits 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 1mo 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 — 117 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Turn a messy working tree (many modified/untracked files spanning several unrelated efforts) into a small number of clean, reviewable commits — each one a single concern — without dragging in excluded topics or unrelated noise.
Hard rule — read this before staging anything
git commit -- <pathspec> does NOT commit the index for those paths. It silently re-stages the full current working-tree content of every matching path first, then commits that.
This means: if you carefully split a file into "wanted hunk" (staged) + "unwanted hunk" (left unstaged) via git apply --cached <patch>, and then run git commit -m "..." -- path/to/that/file, git throws away your careful staging and commits the entire working-tree diff for that file — wanted and unwanted hunks both. This fails silently: the commit succeeds, looks reasonable in the summary, and the pollution only shows up if you diff the commit against what you intended.
Consequence: once ANY file in the commit needs partial-hunk staging (some lines committed now, some left for later/never), you cannot scope that commit with a trailing pathspec. Instead:
- Build the index to contain exactly the target commit's content — nothing more, nothing less:
- Whole files:
git add <path> - Partial files:
git apply --cached <hunk-patch>(see "Splitting a single file" below) — nevergit add <path>on a file you're only partially staging.
- Whole files:
- Check for pre-existing staged content that predates this task (e.g. the user had files already staged before you started —
git statusshows these with a non-space index column). Unstage anything not part of the current commit withgit reset -- <path>(this only touches the index, never the working tree). - Verify the index is exactly right:
git diff --cached --statfor the whole set, andgit diff --cached -- <file>+git diff -- <file>for any partially-staged file (staged half should be the wanted hunk; unstaged remainder should be everything else). - Commit with no trailing pathspec:
git commit -m "...". The index is already correct — trust it. - After every commit, sanity-check it actually contains what you think:
git show <sha> --statand, for any file you partial-staged,git show <sha> -- <file>to eyeball the hunks.
If you skip step 5 and only discover the leak later, see "Recovering from a polluted commit" below — it's fixable without amending.
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.
- 1mo ago First seen · 117 lines · 79 tokens per session scan A 48df6f378fb6
split-commits is a skill published in the GitHub repository tallclub/matimo (18 stars, last pushed yesterday), licensed MIT. It adds 79 tokens to every session and 2,071 once invoked, about $0.0003 per session on Opus 5.5. 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 2026-10-05.
split-commits compared with skills from other repositories
| Mod | Type | Stars | Last push | Licence | Scan | Tokens | Works with |
|---|---|---|---|---|---|---|---|
| split-commitstallclub/matimo | Skill | 18 | yesterday | MIT | A | 79 | Claude Code |
| create-prprisma/orm | Skill | 48k | yesterday | Apache-2.0 | A | 48 | Claude Code, Codex |
| git-commitThinkInAIXYZ/deepchat | Skill | 6.4k | yesterday | Apache-2.0 | A | 15 | Claude Code, Codex |
| repo-syncdavidondrej/skills | Skill | 4.3k | 3d ago | MIT | A | 25 | Claude Code, Codex |
| engineer-bot-opsmakecindy/cindy | Skill | 9.3k | today | Apache-2.0 | A | 57 | Claude Code, Codex |
| mcore-split-prNVIDIA/skills | Skill | 3.9k | yesterday | Apache-2.0 | A | 25 | Claude Code |
| contributingHugoRCD/evlog | Skill | 1.9k | 12d ago | MIT | A | 68 | Claude Code, Codex |