Borrowing it
Nothing to install: this file belongs to dakl/arxiv-mcp. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.
curl -O https://raw.githubusercontent.com/dakl/arxiv-mcp/main/CLAUDE.mdgit clone --depth 1 https://github.com/dakl/arxiv-mcpWrote 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/dakl/arxiv-mcp/claude-md)<a href="https://agentmods.dev/instructions/dakl/arxiv-mcp/claude-md"><img src="https://agentmods.dev/badge/instructions/dakl/arxiv-mcp/claude-md/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/instructions/dakl/arxiv-mcp/claude-md"><img src="https://agentmods.dev/badge/instructions/dakl/arxiv-mcp/claude-md.svg" alt="Reviewed on agentmods" width="80" 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 | $0.00411 | $0.00411 |
| Opus 5.5 | $0.00164 | $0.00164 |
| Sonnet 5.5 | $0.00082 | $0.00082 |
| Haiku 4.5 | $0.00041 | $0.00041 |
Grade A, and why
arxiv-mcp 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 2mo 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
CLAUDE.md
This file provides guidance to Claude Code (claude.ai/code) when working with code in this repository.
What is this?
An MCP (Model Context Protocol) server that exposes arXiv paper search, full-text HTML retrieval (via ar5iv), and PDF download as tools. Built with FastMCP and served over SSE on port 8613.
Commands
uv sync # Install dependencies
uv run arxiv-mcp # Run the server (SSE on 0.0.0.0:8613)
uv run pytest # Run all tests
uv run pytest tests/test_client.py::test_extract_paper_id # Run a single test
uv run ruff check # Lint
uv run ruff format # Format
uv run pyright # Type check (strict mode)
Architecture
src/arxiv_mcp/
domain.py - Pydantic models (Paper)
client.py - arXiv API wrapper (search via `arxiv` lib, HTML via ar5iv, PDF download via httpx)
server.py - FastMCP server exposing three tools: search_papers, get_paper_html, download_pdf
The server is the entrypoint (arxiv-mcp script → server:main). It delegates all arXiv interaction to client.py, which builds Paper domain objects. The search function is synchronous (uses the arxiv library's blocking client); get_html and download_pdf are async (use httpx.AsyncClient).
Conventions
- Python 3.13+, strict pyright, ruff line-length 100
- Pydantic BaseModel for all domain objects in
domain.py - Type hints on everything; avoid
Any - Package managed with uv (never pip)
- Docker:
docker compose up --buildto run containerized
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.
- 2mo ago First seen · 39 lines · 411 tokens per session scan A 8a7101f6f679
arxiv-mcp CLAUDE.md is an instructions file published in the GitHub repository dakl/arxiv-mcp (1 stars, last pushed 9mo ago), licensed MIT. It adds 411 tokens to every session, about $0.0016 per session on Opus 5.5. 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.
arxiv-mcp CLAUDE.md compared with instructions from other repositories
| Mod | Type | Stars | Last push | Licence | Scan | Tokens | Works with |
|---|---|---|---|---|---|---|---|
| arxiv-mcp CLAUDE.mddakl/arxiv-mcp | Instructions file | 1 | 9mo ago | MIT | A | 411 | Claude Code |
| vscode buildNext.instructions.mdmicrosoft/vscode | Instructions file | 195k | yesterday | MIT | A | 9,719 | GitHub Copilot |
| next.js AGENTS.mdvercel/next.js | Instructions file | 143k | yesterday | MIT | A | 7,461 | Codex, OpenCode |
| langchain AGENTS.mdlangchain-ai/langchain | Instructions file | 149k | yesterday | MIT | A | 4,680 | Codex, OpenCode |
| vscode oss-third-party-notices.instructions.mdmicrosoft/vscode | Instructions file | 195k | yesterday | MIT | A | 5,516 | GitHub Copilot |
| vscode copilot-instructions.mdmicrosoft/vscode | Instructions file | 195k | yesterday | MIT | A | 3,093 | GitHub Copilot |
| deepseek-harness AGENTS.mddeepseek-ai/deepseek-harness | Instructions file | 274k | 2d ago | MIT | A | 4,313 | Codex, OpenCode |