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/mpecan/tokf/claude-mdgit clone --depth 1 https://github.com/mpecan/tokfWrote 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/instructions/mpecan/tokf/claude-md)<a href="https://agentmods.dev/instructions/mpecan/tokf/claude-md"><img src="https://agentmods.dev/badge/instructions/mpecan/tokf/claude-md.svg" alt="Measured on agentmods" 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 | $0.02341 | $0.02341 |
| Opus 5 | $0.01171 | $0.01171 |
| Sonnet 5 | $0.00468 | $0.00468 |
| Haiku 4.5 | $0.00234 | $0.00234 |
Grade A, and why
tokf 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 4d 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 — 205 lines — stays where its author put it; the contents beside it link to each section on GitHub.
tokf — Development Guidelines
Project Philosophy
tokf is an open source project built for the community. We are not looking for profits — this exists for open source sake. Every decision should prioritize:
- End-user experience — whether the user is a human or an LLM, the tool should be intuitive, fast, and transparent about what it's doing.
- Visibility — users should always understand what tokf is doing. Stderr notes,
--timing,--verboseflags. Never hide behavior. - Transparency — clear error messages, honest documentation, no dark patterns.
Commits
Use Conventional Commits strictly:
<type>(<scope>): <description>
[optional body]
[optional footer]
Types: feat, fix, refactor, test, docs, chore, ci, perf, build
Scopes: config, filter, runner, output, cli, hook, tracking, history
Examples:
feat(filter): implement skip/keep line filteringfix(config): handle missing optional fields in git-status.tomltest(filter): add fixtures for cargo-test failure caseci: add clippy and fmt checks to GitHub Actions
Keep commits atomic — one logical change per commit. Don't bundle unrelated changes.
Code Quality
Testing
- Minimum 80% coverage, target 90%.
- Every module gets unit tests. Every filter gets integration tests with fixture data.
- Fixture-driven: save real command outputs as
.txtfiles intests/fixtures/. Tests load fixtures, apply filters, assert on output. No dependency on external tools in tests. - Declarative filter tests: place test cases in a
<stem>_test/directory adjacent to the filter TOML (e.g.filters/git/push_test/next tofilters/git/push.toml). Each case is a TOML file withname,inlineorfixture,exit_code, and[[expect]]blocks. Run withtokf verify. Every filter in the stdlib must have a_test/suite — CI enforces this withtokf verify --require-all. - Run
cargo testafter every meaningful change. Tests must pass before committing.
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.
- 4d ago First seen · 205 lines · 2,341 tokens per session scan A 015df1a4f3e2
tokf CLAUDE.md is an instructions file published in the GitHub repository mpecan/tokf (196 stars, last pushed yesterday), licensed MIT. It adds 2,341 tokens to every session, about $0.0117 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 2026-08-30.
Other instructions, from other repositories
scraper-mcp CLAUDE.md
Instructions for cotdp/scraper-mcp, covering claude.md, project overview, development commands, environment setup and install dependencies (uses uv package manager).
ClaudeShrink CLAUDE.md
Claude Code instructions for g-akshay/ClaudeShrink, covering claudeshrink, skill, installation, usage and development.
ai-workflow AGENTS.md
Instructions for cunhaax/ai-workflow, covering ai workflow template, rules — non-negotiable, project overview, commands and architecture.
vibetags com-example-config-AsyncTestConfig.instructions.md
Instructions for PIsberg/vibetags, covering copilot instructions for asynctestconfig and immutable type.
pupila CLAUDE.md
Instructions for FranRom/pupila, covering claude.md, where things live, stack, run locally and repo layout.
claude-bridge AGENTS.md
Instructions for kadaba/claude-bridge, covering agents.md — using claude-bridge from any ai coding cli, two independent channels, 1. coordination (the conversation), 2. file transfer (the bytes) — same for every tool and workflows.