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 instructions/neturely/okffs/claude-mdgit clone --depth 1 https://github.com/neturely/okffsWhat 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 | $0.11173 | $0.11173 |
| Opus 5 | $0.05587 | $0.05587 |
| Sonnet 5 | $0.02235 | $0.02235 |
| Haiku 4.5 | $0.01117 | $0.01117 |
Grade A, and why
okffs CLAUDE.md 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 — 181 lines — stays where its author put it; the contents beside it link to each section on GitHub.
CLAUDE.md
Guidance for Claude Code when working in this repository.
Project
okffs is a TypeScript/Node.js MCP server that connects Claude Code to GitHub, enabling a full issue → branch → merge → close workflow. The goal: discuss tasks in Claude.ai, then push them to GitHub as issues and branches in one shot via Claude Code.
Stack
- TypeScript / Node.js MCP server
- GitHub Personal Access Token (PAT) authentication (upgrade to a GitHub App later)
- Published to npm (package name:
@neturely/okffs) and the MCP Registry (registry.modelcontextprotocol.io)
Conventions
- Destructive tools require
confirmed: true— call once for a warning, re-call to proceed. - GitHub is the source of truth for issue state, never local.
- Keep the tool surface minimal — do one thing well per tool.
- Destructive actions post a comment to the issue before acting.
- Always reach for an okffs tool before raw
git/gh/GraphQL when one covers the action — a correctness rule, not a style preference. okffs authenticates with the configuredGITHUB_TOKEN(scoped for this repo, incl. Projects/org Issue Fields), whereas rawgh/gituses whatever ambient CLI token is signed in — often the wrong identity or missing scopes; and okffs applies the conventions (branch naming, the Branch: link,Closes #N, board placement, changelog fragments, theOKFFS_PROTECTED_BRANCHinvariant) that hand-rolled commands silently skip. When an okffs tool covers the action — issues, PRs (create_pull_requestfor an issue's PR;promote_branchfor a base→protected promotion likedevelop → main, never rawgh pr create;fix_into_basefor an issue-less fix PR into the base branch, never rawgh pr create/gh pr merge), comments, review threads (resolve_review_thread/ theaddress_pr_reviewprompt), releases (prepare_release), project status — use it and honour its env toggles (OKFFS_RESOLVE_THREADS,OKFFS_BASE_BRANCH,OKFFS_PROTECTED_BRANCH,OKFFS_PROMOTION_*, …) rather than re-deriving the behaviour. Fall back to rawgit/ghonly when no okffs tool fits. Never merge/tag/publish intoOKFFS_PROTECTED_BRANCHautonomously (okffs may open a PR into it, but not merge) — hand back to the user.
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 · 181 lines · 11,173 tokens per session scan A 2c289721c0f7
okffs CLAUDE.md is an instructions file published in the GitHub repository neturely/okffs (2 stars, last pushed 12d ago), licensed MIT. It adds 11,173 tokens to every session, about $0.0559 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 instructions, from other repositories
koan AGENTS.md
AGENTS.md instructions for Anantys-oss/koan, covering agents.md, project overview, documentation first, wiki tooling and commands.
koan CLAUDE.md
Claude Code instructions for Anantys-oss/koan, covering claude.md, what is kōan, on-demand guidance (nested claude.md), specs discipline (mandatory) and documentation first.
koan copilot-instructions.md
Copilot instructions for Anantys-oss/koan, a project described as: Autonomous background agent that consumes unused Claude Max quota to explore, audit, and improve a codebase — communicating with its human via Telegram and a shared mission queue.
claude-codex-settings CLAUDE.md
Claude Code instructions for fcakyon/claude-codex-settings, covering claude-settings, repo structure, cross-tool plugin format reference, plugin manifests and root marketplace files.
vscode-gitlens CLAUDE.md
Claude Code instructions for gitkraken/vscode-gitlens, a project described as: Supercharge Git inside VS Code and unlock untapped knowledge within each repository — Visualize code authorship at a glance via Git blame annotations and CodeLens, seamlessly navigate and explore Git repositories, gain valuable insights via…
ghtkn AGENTS.md
Instructions for suzuki-shunsuke/ghtkn, a project described as: A CLI to create short-lived (8 hours) GitHub App User Access Token for secure local development.