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/nel-neru/libraium/agents-mdgit clone --depth 1 https://github.com/nel-neru/LibrAIumWhat 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 | $0.03264 | $0.03264 |
| Opus 5 | $0.01632 | $0.01632 |
| Sonnet 5 | $0.00653 | $0.00653 |
| Haiku 4.5 | $0.00326 | $0.00326 |
Grade A, and why
LibrAIum 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.
Copies of this mod
1 near-identical copy found in the catalogue:
- LibrAIum CLAUDE.md — 94% identical, 12 lines differ
How it starts
The opening of the file, as written. The whole thing — 81 lines — stays where its author put it; the contents beside it link to each section on GitHub.
AGENTS.md
This file provides guidance to Codex (Codex.ai/code) when working with code in this repository.
Language
Always converse with the user in Japanese — replies, explanations, plans, questions, and progress reports. Keep repository artifacts in English: code, comments, commit messages, README/docs, and identifiers (the repo is published publicly in English; the owner reads both).
What LibrAIum Is
A local-first Tauri v2 + Svelte 5 desktop app for curating best-practice GitHub repositories, storing everything as YAML-frontmatter Markdown files in a local git repository, and exposing the library to Codex through an MCP server. The authoritative spec is LibrAIum_完全設計書_v1.0.md (Japanese). v1.0 MVP is fully implemented.
Commands
npm run tauri dev # run the desktop app (Rust compiles on first run)
npm run tauri build # package release build
npm run build # frontend-only production build (vite)
npm test # frontend unit tests (markdown-renderer hardening, node --test)
cd src-tauri && cargo test # Rust unit tests — all core logic lives here
cd src-tauri && cargo test store # run one module's tests (store/search/gitops/github/…)
cd mcp-server && npm test # store.js unit tests + MCP stdio smoke test (all 8 tools)
bash scripts/make-icons.sh # regenerate icons from scripts/generate-icons.mjs (macOS)
node scripts/register-mcp.mjs # user-scope MCP registration plan (--yes to apply, --doctor to diagnose)
Rust was installed via Homebrew; if cargo is missing from PATH: export PATH="/opt/homebrew/bin:$PATH".
Note: cargo test/cargo build require dist/ to exist (Tauri's generate_context! embeds it) — run npm run build first on a fresh clone.
Architecture
Two independent consumers share one data format:
- Desktop app — Svelte 5 GUI (
src/) → Tauri commands (src-tauri/src/commands.rs) → Rust core modules. - MCP server (
mcp-server/, Node stdio) — reads/writes the samedata/directory directly.
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 · 81 lines · 3,264 tokens per session scan A 3bb4eda934ad
LibrAIum AGENTS.md is an instructions file published in the GitHub repository nel-neru/LibrAIum (0 stars, last pushed 6d ago), licensed MIT. It adds 3,264 tokens to every session, about $0.0163 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
rag-rat AGENTS.md
Instructions for cq27-dev/rag-rat, covering agents.md, prefer the rag-rat mcp for code browsing, record durable learnings as rag-rat memories, repo orientation and worktree correctness.
codex-image-context-runtime AGENTS.md
Instructions for shixinnt/codex-image-context-runtime, covering agents.md, public boundary, runtime contract and changes.
blz AGENTS.md
Instructions for outfitter-dev/blz, covering blz repository instructions for ai agents, important, working memory, use blz and 🚀 quick start for agents.
Khaos-Brain AGENTS.md
Instructions for liuyingxuvka/Khaos-Brain, covering repository expectations, bootstrap, start here, purpose and github publish default.
memem CLAUDE.md
Claude Code instructions for TT-Wang/memem, covering memem — persistent memory & context assembly, auto-recall, context model (v2.8+), tier 1 — profiles (always-injected) and tier 2 — working rules (procedural, citation-ranked).
google-ads-mcp-unofficial AGENTS.md
Instructions for davidmosiah/google-ads-mcp-unofficial, covering agent-facing usage — google-ads-mcp-unofficial, tool count: 22, agent first-call ladder, hard rules (do not violate) and money safety.