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/fmflurry/settings-opencodeWrote 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/rules/fmflurry/settings-opencode/git-workflow)<a href="https://agentmods.dev/rules/fmflurry/settings-opencode/git-workflow"><img src="https://agentmods.dev/badge/rules/fmflurry/settings-opencode/git-workflow/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/rules/fmflurry/settings-opencode/git-workflow"><img src="https://agentmods.dev/badge/rules/fmflurry/settings-opencode/git-workflow.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 | $0.01428 | $0.01428 |
| Opus 5.5 | $0.00571 | $0.00571 |
| Sonnet 5.5 | $0.00286 | $0.00286 |
| Haiku 4.5 | $0.00143 | $0.00143 |
Grade B, and why
git-workflow scanned grade B with 1 finding 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 8d 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.
Reads agent configuration directoriesmediumAgent snooping
.claude/, .codex/, .gemini/ hold keys, settings and other credentials a mod has no legitimate need for.
Note: Attribution disabled globally via ~/.claude/settings.json. How it starts
The opening of the file, as written. The whole thing — 132 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Git Workflow
Commit Message Format
<type>: <description>
<optional body>
Types: feat, fix, refactor, docs, test, chore, perf, ci
Note: Attribution disabled globally via ~/.claude/settings.json.
Language (mandatory)
Every commit message MUST be written in English — subject, body, and footers — regardless of the user's language, the orchestrator's language, or the conversation language. French (or any other non-English) commit messages are a hard violation; do not commit until rewritten. When the user describes a change in French, translate the intent into an English summary — never quote or transliterate the French. Conventional Commits format and the type/scope/length rules above are unchanged. The same English requirement applies to PR titles and PR bodies.
Precedence (mandatory)
This is the canonical spec — git-specialist operates against these exact rules. Do not restate the full spec elsewhere; reference this section instead.
Logical cohesion is the hard rule; commit size is a signal, never a blocker. A commit is one coherent logical set of changes. Size never justifies breaking a coherent set apart. When cohesion and a size number conflict, cohesion wins.
Counting Scope
The size signal below counts only review-weight changes: application source code and application configuration.
Exempt from the count: test fixtures and test data (Bruno .bru, .http, seed files, test JSON/CSV/XML), snapshots, documentation and markdown, generated migrations, generated/scaffolded code, lock files (package-lock.json, yarn.lock, pnpm-lock.yaml, bun.lock), binaries/assets, pure renames detected as R100.
Commit Size Signal (non-blocking)
More than ~15 counted files or ~400 counted lines in a commit → look for a legal split (see Legal-Split Test below) and report the observed counted numbers. This is a signal to check for a natural split — never a reason to refuse, block, unstage, or delay a commit.
Legal-Split Test
A split is legal only if every resulting commit has:
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.
- 8d ago First seen · 132 lines · 1,428 tokens per session scan B 8a645e004976
git-workflow is a cursor rule published in the GitHub repository fmflurry/settings-opencode (171 stars, last pushed yesterday), licensed MIT. It adds 1,428 tokens to every session, about $0.0057 per session on Opus 5.5. A static security scan graded it B with 1 finding (reads agent configuration directories). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-10-08.
Other cursor rules, from other repositories
git
Git workflow — commit format, push policy, and PR process.
commit-sponsors
Every git commit message must acknowledge BetterDesk supporters.
no-ai-contributors
Never add AI tools as git authors or project contributors.
git-discipline
Commit, push, and branching rules. Owns ALL git behavior; other files state only their deviations.
git-commit-standards
Enforce structured Git commit messages.
conventional-commits
Use the Conventional Commit Messages specification to generate commit messages.