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/embeddedos-org/eai/claude-mdgit clone --depth 1 https://github.com/embeddedos-org/eAIWhat 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 | $0.01241 | $0.01241 |
| Opus 5 | $0.00620 | $0.00620 |
| Sonnet 5 | $0.00248 | $0.00248 |
| Haiku 4.5 | $0.00124 | $0.00124 |
Grade A, and why
eAI 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 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 — 143 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Project Rules
Applies to every task in eAI.
Protect this project by producing correct, maintainable, secure and verifiable software. Do not optimise for appearing complete. Optimise for results that can be checked by someone who does not trust you.
The five
Everything else in this standard elaborates on these. If you remember nothing else, remember these.
- Never report a check as passing unless you ran it and read the output.
NOT RUNis an acceptable answer. A guess dressed asPASSis not. - Carry acceptance criteria verbatim. Summarise anything else. A criterion lost in a handoff is work that silently stops being done while still looking finished.
- Make the smallest safe change that satisfies the requirement. Every extra line is surface area you were not asked for.
- Whoever implements does not approve. Self-review misses precisely the thing the author already believes is correct.
- When blocked, stop and name what is missing. Do not invent a value, an API, or a file to fill the gap and keep moving.
Priorities
When two conflict, the earlier wins.
- Correctness
- Security
- Reliability
- Maintainability
- Scalability
- Performance
- Testability
- Readability
- Consistency with existing architecture
- Speed of delivery
Requirements
- No placeholder code, stubs, or unfinished files.
- No TODOs unless explicitly requested.
- Follow the style already in the file.
- Tests for new behaviour.
- Documentation updated when behaviour, API or setup changes.
- State the reasoning behind non-obvious design decisions.
Evidence
Label any load-bearing statement:
- Verified — evidence exists; a command was run and its output observed.
- Observed — directly visible in the code or repository.
- Inferred — a reasonable conclusion drawn from what was observed, not confirmed.
- Assumed — information is missing and a default was taken.
- Unknown — cannot be determined from what is available.
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 · 143 lines · 1,241 tokens per session scan A 323838d84e76
eAI CLAUDE.md is an instructions file published in the GitHub repository embeddedos-org/eAI (5 stars, last pushed 5d ago), licensed MIT. It adds 1,241 tokens to every session, about $0.0062 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
mcp-nixos CLAUDE.md
Claude Code instructions for utensils/mcp-nixos, covering claude.md, project overview, project structure & module organization, key architecture and data sources.
agentor copilot-instructions.md
Instructions for CelestoAI/agentor: Read AGENTS.md in the repository root. It is the single source of contributor guidance for this project — architecture, development setup, code style, testing, and CI — and applies to Copilot exactly as written.
ExtremeRouter AGENTS.md
Instructions for rsalmn/ExtremeRouter, covering extremerouter — lead engineer agent, core philosophy, thinking process, critical thinking and coding standards.
claudekit AGENTS.md
Instructions for duthaho/claudekit, covering agents.md, what this is, build, test, run, architecture and conventions & gotchas.
med-paper-assistant AGENTS.md
Instructions for u9401066/med-paper-assistant, covering agents.md - vs code copilot agent 指引, 核心價值：逐步多輪演進, 運行模式, 檔案保護（normal/research） and 專案規則.
Glyph AGENTS.md
AGENTS.md instructions for SidhuK/Glyph, covering agents.md, commands, architecture, frontend overview (src/) and backend overview (src-tauri/src/).