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/permacommons/chabeau/agents-mdgit clone --depth 1 https://github.com/permacommons/chabeauWhat 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 | $0.00506 | $0.00506 |
| Opus 5 | $0.00253 | $0.00253 |
| Sonnet 5 | $0.00101 | $0.00101 |
| Haiku 4.5 | $0.00051 | $0.00051 |
Grade A, and why
chabeau AGENTS.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 — 28 lines — stays where its author put it; the contents beside it link to each section on GitHub.
This repository contains Chabeau, an open source (CC-0) chatbot TUI for remote API-based access.
General guidelines
- Follow best practices for Rust development and build/test/release processes.
- Ensure new functionality does not introduce usability, security or accessibility regressions.
- Add unit tests opportunistically.
- Update docs as needed after you've completed the main objective:
- Put user-facing behavior, usage guidance, and repository file-tree updates in
README.md. - Put architecture notes, internal module responsibilities, and design rationale in
ARCHITECTURE.md.
- Put user-facing behavior, usage guidance, and repository file-tree updates in
- Always run
cargo test,cargo check,cargo fmtandcargo clippyafter you've completed the main objective. - Feel free to suggest changes to WISHLIST.md after all these steps have been completed. Thanks for making Chabeau better!
- Do not add bugfix notes to the README. Keep README focused on user-facing features and behavior; document fixes in commit messages.
Additional style note for agents
- Do not leave breadcrumb comments about past changes (e.g., "removed", "replaced", or historical notes in code). Keep comments focused on current behavior and intent only.
Commit summaries
- When asked to draft a commit message, use a conventional commit format. The first line should be 50 characters wide max; subsequent lines 72 characters wide max. Keep it under 5 bullet points. Only summarize uncommitted (staged or working area) changes, not prior versions of those changes (e.g., if you refactored a function, and then refactored it again, only summarize its final state)
- When constructing commit messages via CLI flags, prefer a single
-mbody with newline-separated bullets (or a heredoc) rather than multiple-mflags, to avoid extra blank lines between bullets. - When in interactive mode, present commit summaries to the user in a way that is easy to copy and paste with any formatting (e.g., markdown) intact.
- Do not mention passing tests.
Performance validations
- When you implement performance-sensitive features (rendering, scrolling, parsing, streaming), validate with Criterion benchmarks in addition to unit tests.
- See
benches/README.mdfor a quick scaffold on adding benches locally (and how to export internal modules temporarily viasrc/lib.rs). - Keep unit-test perf checks green (
cargo test). Use benches to quantify improvements and regressions (cargo bench).
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 · 28 lines · 506 tokens per session scan A 140deebd0c28
chabeau AGENTS.md is an instructions file published in the GitHub repository permacommons/chabeau (40 stars, last pushed 1mo ago), licensed CC0-1.0. It adds 506 tokens to every session, about $0.0025 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
tuicr AGENTS.md
AGENTS.md instructions for agavra/tuicr, covering agents.md - tuicr, project structure, key types, data flow and important implementation details.
Monoloth AGENTS.md
Instructions for noahain/Monoloth, covering agents.md, stack, layout, build / test commands and from repo root.
oneclaw CLAUDE.md
Instructions for 0xNyk/oneclaw, covering claude.md, build & development commands, architecture, core traits (the extension points) and adding a new implementation.
codex-session-tui AGENTS.md
Instructions for avikalpa/codex-session-tui, covering agents.md, project purpose, current scope, architecture and safety rules.
r3bl-open-core GEMINI.md
Instructions for r3bl-org/r3bl-open-core, a project described as: TUI framework and developer productivity apps in Rust 🦀.
cli AGENTS.md
Instructions for googleworkspace/cli, covering agents.md, project overview, build & test, changesets and architecture.