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/arcadeai-labs/agent-library/claude-mdgit clone --depth 1 https://github.com/arcadeai-labs/agent-libraryWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/instructions/arcadeai-labs/agent-library/claude-md)<a href="https://agentmods.dev/instructions/arcadeai-labs/agent-library/claude-md"><img src="https://agentmods.dev/badge/instructions/arcadeai-labs/agent-library/claude-md.svg" alt="Measured on agentmods" height="20"></a>What 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 5.1 | $0.04612 | $0.04612 |
| Opus 5 | $0.02306 | $0.02306 |
| Sonnet 5 | $0.00922 | $0.00922 |
| Haiku 4.5 | $0.00461 | $0.00461 |
Grade A, and why
agent-library 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 6d 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 — 527 lines — stays where its author put it; the contents beside it link to each section on GitHub.
CLAUDE.md
This file provides guidance to Claude Code (claude.ai/code) when working with code in this repository.
Project Overview
Agent Library (package name: agent-library, Python module: librarian) is a multi-modal knowledge library for AI agents built on Arcade for the Model Context Protocol (MCP). It provides persistent storage with semantic and keyword search for text, code, images, and PDFs.
Naming note for contributors: The product is "Agent Library." The Python module path is librarian (don't rename — it's the import surface). The MCP server identifies itself as Librarian (MCPApp(name="Librarian") in server.py), which is why exposed tools carry the Librarian_ prefix. Keep that distinction in mind: prose says "Agent Library"; code-level identifiers say "librarian"/"Librarian".
Key Technologies
- SQLite with
sqlite-vecfor vector search - FTS5 with BM25 ranking for full-text search
- Hybrid search combining both approaches with Max Marginal Relevance (MMR)
- Configurable embedding models (local sentence-transformers or OpenAI-compatible API)
- Support for multi-modal assets (text, code, images, PDFs)
Current Status
- Multi-Modal Support Complete: Code, PDF, and image parsing with asset type preservation
- Parser Registry: Automatic parser selection based on file extension
- Database Schema: Multi-modal columns (asset_type, modality_data) fully implemented
- Search Integration: All search tools return asset_type to AI agents
- See
IMPLEMENTATION_STATUS.mdfor detailed progress tracking - See
MULTI_MODAL_LIBRARIAN_DESIGN.mdfor complete design specification
Development Commands
Setup & Installation
./setup.sh # Initial setup
make install # Install base dependencies
make sync # Sync dependencies from pyproject.toml
# Install optional multi-modal dependencies
uv pip install -e ".[pdf]" # PDF processing (pypdf)
uv pip install -e ".[vision]" # Image processing (Pillow)
uv pip install -e ".[all]" # All multi-modal features
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.
- 6d ago First seen · 527 lines · 4,612 tokens per session scan A ed7d8e0ad540
agent-library CLAUDE.md is an instructions file published in the GitHub repository arcadeai-labs/agent-library (32 stars, last pushed 1mo ago), licensed Apache-2.0. It adds 4,612 tokens to every session, about $0.0231 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
trueforge AGENTS.md
AGENTS.md instructions for truefoundry/trueforge, a project described as: The open-source agent harness - the runtime layer that turns an LLM into a working agent.
relay AGENTS.md
AGENTS.md instructions for AgentWorkforce/relay, covering git workflow rules, never push directly to main, correct workflow, ... do work .. and stop here - let user merge.
puppyone AGENTS.md
Instructions for puppyone-ai/puppyone, covering puppyone (contextbase), connect, collaborate, platform and active development directories.
LeAgent AGENTS.md
Instructions for vixues/LeAgent, covering leagent development guidelines, project overview, architecture, execution topology (one kernel, many ingresses) and surface map (where things live).
Ultimate-Agent-Directory AGENTS.md
AGENTS.md instructions for moshehbenavraham/Ultimate-Agent-Directory, covering agents.md, important rule, repository overview, essential build commands and data architecture.
grix AGENTS.md
AGENTS.md instructions for askie/grix, covering grix agent guide, repository boundaries, required workflows, shared agent configuration and cross-component contracts.