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/utensils/claudette/rustgit clone --depth 1 https://github.com/utensils/claudetteWhat 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 | $0.00498 | $0.00498 |
| Opus 5 | $0.00249 | $0.00249 |
| Sonnet 5 | $0.00100 | $0.00100 |
| Haiku 4.5 | $0.00050 | $0.00050 |
Grade A, and why
claudette rust.instructions.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.
What it actually says
Rust code uses edition 2024, default rustfmt, and clippy with warnings denied in CI. Prefer clear ownership, explicit error handling, and small focused modules over large multi-purpose functions.
Keep crate boundaries intact. Business logic belongs in the claudette crate under src/. Tauri command files should parse inputs, open state/resources, call core services, and return serializable results. Do not move durable logic into the webview or into Tauri wrappers just because the caller is UI-driven.
Model types belong in src/model/, should be IO-free, and should derive Serialize. Database code should open fresh rusqlite connections per command because Connection is not Send.
For migrations, create a new timestamped SQL file and add a Migration entry. Do not modify released migration SQL. Prefer additive, forward-only migrations and include IF NOT EXISTS where appropriate.
For process, git, terminal, MCP, plugin, workspace, and agent changes, preserve current lifecycle semantics unless the task explicitly changes them: cancellation, cleanup, event ordering, worktree branch state, persisted session state, and streaming message chronology are regression-sensitive.
When adding backend behavior, add focused tests close to the changed module. Use tempfile::tempdir() for git repos and transient DBs. Use #[tokio::test] for async behavior.
Avoid expanding god files. If adding a separate concern to src/diff.rs, src/git.rs, src/plugin.rs, src/mcp.rs, src/mcp_supervisor.rs, src-tauri/src/ipc.rs, src-tauri/src/voice.rs, or a large command file, extract a helper module or domain file and keep the entry point as orchestration.
Gate OS-specific code with #[cfg(windows)], #[cfg(unix)], #[cfg(target_os = "macos")], or matching negative cfgs. Do not assume Unix paths or shells in cross-platform code. Never spawn a subprocess with a raw std::process::Command::new / tokio::process::Command::new — on Windows that flashes a blank cmd.exe console window in release builds. Use the claudette::process helper or CommandWindowExt::no_console_window(); use .new_console_window() only for a terminal the user intentionally opens.
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 · 20 lines · 498 tokens per session scan A 1b3734a9b4f6
claudette rust.instructions.md is an instructions file published in the GitHub repository utensils/claudette (75 stars, last pushed 2d ago), licensed MIT. It adds 498 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
kungfu copilot-instructions.md
Copilot instructions for kungfu-systems/kungfu: Read and follow AGENTS.md before proposing or running changes. It is the shared agent router for this repository, including the canonical Shifu development and build entrypoint.
sofagent GEMINI.md
Instructions for KongFangXun/sofagent, covering gemini.md · sofagent gemini cli 适配（薄挂载）, 会话开始时按序加载（四层加载链）, 审计强制（平台无关） and 连接 mcp server.
agents CLAUDE.md
Instructions for eloylp/agents, covering claude.md, project overview, directory structure, config model and build & run.
ideate AGENTS.md
Instructions for paultyng/ideate, covering agents.md, build features as mcp tools first, ui second, when to add ui and pattern.
growmos AGENTS.md
Instructions for codician-team/growmos: This repository keeps a knowledge graph in .growmos/ (entities, typed relations, provenance, profiles, a journal). It is the shared world model that survives context windows. Treat it as memory you read at the start of work and write to as you develop. Zero-config commands.
omk CLAUDE.md
Claude Code instructions for dmae97/omk, covering claude.md, precedence (read this first), omk runtime map (what you're operating inside), this repo: open-multi-agent-kit / omk monorepo and build / test / check commands.