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/salam/claudine/agents-mdgit clone --depth 1 https://github.com/salam/claudineWhat 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 | $0.00723 | $0.00723 |
| Opus 5 | $0.00362 | $0.00362 |
| Sonnet 5 | $0.00145 | $0.00145 |
| Haiku 4.5 | $0.00072 | $0.00072 |
Grade A, and why
claudine 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 3d 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 — 52 lines — stays where its author put it; the contents beside it link to each section on GitHub.
- When interacting with the user, print ▸ before answering anything else.
- Keep the project folder clean.
- Never commit secrets.
- Never touch, edit or move CLAUDE.md, AGENT.md and AGENTS.md files.
- Place shell scripts, if not necessary otherwise, in the folder ./tools/
- Create a ./RELEASE_NOTES.md file for each new feature implemented. Use user-targeted language, short and brief bullet points. Create a new section ## Release 1.2 (Mon, Jan 19 09:39) for every 4 hours. E.g.
Version 1.0.3 (Feb 7 2025, 17:53)
- Feature 1
- Feature 2
-
Always compile and run tests before considering a prompted task as done.
-
Keep a .GIT_NEXT_COMMIT_MESSAGE.md file that contains a summary and diff of staged changes since the last commit and is used to prefill the commit message.
-
Use plain english when explaining (technical) concepts, and still refer to the professional computer science terminology in brackets.
-
Before attempting to delete a file to resolve a local type/lint failure, stop and ask the user. Other agents are often editing adjacent files; deleting their work to silence an error is never acceptable without explicit approval.
-
NEVER edit .env or any environment variable files—only the user may change them.
-
Coordinate with other agents before removing their in-progress edits—don't revert or delete work you didn't author unless everyone agrees.
-
Moving/renaming and restoring files is allowed.
ATOMIC COMMITS
- Keep commits atomic: commit only the files you touched and list each path explicitly. For tracked files run
git commit -m "<scoped message>" -- path/to/file1 path/to/file2. For brand-new files, use the one-linergit restore --staged :/ && git add "path/to/file1" "path/to/file2" && git commit -m "<scoped message>" -- path/to/file1 path/to/file2
SPECS
- When a new specification is given, check for contradictions and highlight them to the prompting user.
- If a given specification is incomplete, ask for clarifications if necessary.
- For a given specification, write a short todo list in the FEATURES.md file.
- For a given specification, write a unit test that covers different scenarios.
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.
- 3d ago First seen · 52 lines · 723 tokens per session scan A 6871079e84fd
claudine AGENTS.md is an instructions file published in the GitHub repository salam/claudine (21 stars, last pushed 4mo ago), licensed MIT. It adds 723 tokens to every session, about $0.0036 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
collattice CLAUDE.md
Instructions for MrBildo/collattice, covering collattice, repository rules — hard, tech stack, build & run and prerequisites.
kagan AGENTS.md
Instructions for kagan-sh/kagan, covering agents.md, commands, build and package, code map and architecture constraints.
sanban AGENTS.md
AGENTS.md instructions for aancw/sanban, covering agents.md — guidance for ai agents, project, file map, api base and common tasks.
kilocode AGENTS.md
AGENTS.md instructions for Kilo-Org/kilocode, covering agents.md, build and dev, quality checks, products and package instructions.
onedev AGENTS.md
AGENTS.md instructions for theonedev/onedev: Do not add Co-authored-by: or other co-author / AI attribution footers to commit messages unless the user explicitly asks for them. If such a trailer is injected into a commit you just created (and it is not pushed), rewrite the commit message to remove it before finishing.
veritas-kanban AGENTS.md
Instructions for BradGroux/veritas-kanban, covering agents.md — canonical agent instructions for veritas kanban, runtime requirements, repository layout, essential commands and install.