What is cochrane-mcp CLAUDE.md?

Developer instructions for a server that searches the Cochrane Library, a collection of medical evidence reviews and clinical trial records. They explain the server's tools, data types, and a browser-related access requirement.

Use them when developing or operating the Cochrane search server, especially when searching reviews, retrieving trial details, or diagnosing access failures.

They help an AI agent use the server correctly and account for the website's Cloudflare security check, which may require a real Chrome browser.

How aliildan/cochrane-mcp configures Claude Code

Reuse

Borrowing it

Nothing to install: this file belongs to aliildan/cochrane-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.

Copy the file

curl -O https://raw.githubusercontent.com/aliildan/cochrane-mcp/main/CLAUDE.md

Clone the repo

git clone --depth 1 https://github.com/aliildan/cochrane-mcp

Made for: Claude Code.

Wrote 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.

README.md

[![agentmods](https://agentmods.dev/badge/instructions/aliildan/cochrane-mcp/claude-md/github.svg)](https://agentmods.dev/instructions/aliildan/cochrane-mcp/claude-md)

Your own site

<a href="https://agentmods.dev/instructions/aliildan/cochrane-mcp/claude-md"><img src="https://agentmods.dev/badge/instructions/aliildan/cochrane-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.

Your own site · 80×15

<a href="https://agentmods.dev/instructions/aliildan/cochrane-mcp/claude-md"><img src="https://agentmods.dev/badge/instructions/aliildan/cochrane-mcp/claude-md.svg" alt="Reviewed on agentmods" width="80" height="20"></a>

Per session 2,384 This file is loaded in full into every session.

When invoked 2,384 The same file — it is already loaded in full.

Security scan C 2 findings. A grade says what 26 rules found in the file — not that it is safe.

Origin original No closer match found in the catalogue.

Token cost

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	$0.02384	$0.02384
Opus 5	$0.01192	$0.01192
Sonnet 5	$0.00477	$0.00477
Haiku 4.5	$0.00238	$0.00238

Measured 8d ago against content hash ef8da8ddfdbe, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-08, from the pricing page.

Security

Grade C, and why

cochrane-mcp CLAUDE.md scanned grade C with 2 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 8d 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.

Recursive force deletehighDestructive command

rm -rf with a variable or a broad path is one typo away from removing the wrong tree.

To force a refresh of a stuck npx copy: `rm -rf ~/.npm/_npx/*/node_modules/cochrane-mcp` (or the whole

Makes network callslowCapability

Not a fault in itself. Listed so you know the mod talks to something, and to what.

- Plain `fetch` / `curl` / TLS-impersonation (`impit`) → **HTTP 412** once the IP is flagged.

CLAUDE.md · 153 lines

How it starts

The opening of the file, as written. The whole thing — 153 lines — stays where its author put it; the contents beside it link to each section on GitHub.

CLAUDE.md

Guidance for Claude Code (and other AI agents) working in this repository.

What this is

cochrane-mcp is a stdio MCP server (TypeScript, ESM) that searches the Cochrane Library and returns structured results and rich, type-aware details. It is also packaged as a Claude Code plugin.

Tools exposed:

cochrane_search — search any content type; returns total, per-type typeCounts, and a page of items.
cochrane_get_details — full record for a DOI (reviews: metadata + structured abstract + plain-language summary + PICO + related; trials: metadata + source registry).
cochrane_suggest_terms — autocomplete term suggestions.

⚠️ The one thing you must understand: Cloudflare

The Cochrane Library is behind a Cloudflare JS/managed challenge. This is the defining constraint of the whole codebase. Empirically:

Plain fetch / curl / TLS-impersonation (impit) → HTTP 412 once the IP is flagged.
A freshly automation-launched headless or headful Chrome also fails to clear the challenge.
A real, already-running Chrome (or one where a human solves the one-time Turnstile) clears it and mints a cf_clearance cookie.

So the engine is mint-then-fetch:

CdpMinter obtains cf_clearance from a real browser, choosing automatically: explicit COCHRANE_CDP_ENDPOINT → discover a running debug Chrome on 127.0.0.1:9222/9444 (reused only if it already holds clearance; connecting is read-only and never closes the user's Chrome) → self-launch a persistent-profile Chrome (system Chrome, else bundled Chromium).
HttpClient replays that cookie + the browser's User-Agent in fast plain fetch calls.
On a 412/challenge it refreshes the cookie once and retries.

Redirect handling matters: Cochrane (Liferay/Atypon) answers detail pages with a 302 → same URL that sets a fresh JSESSIONID/SID_REP/SCOLAUTHSESSIONID. HttpClient.fetchFollowing therefore follows redirects manually, merging each hop's Set-Cookie into the jar (like a browser) — using fetch's automatic redirect (fixed Cookie header) loops forever (redirect count exceeded). Don't revert this to redirect: "follow".

Read the full file on GitHub · 153 lines

Changes

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.

8d ago First seen · 153 lines · 2,384 tokens per session scan C ef8da8ddfdbe

Subscribe to this mod's changes

cochrane-mcp CLAUDE.md is an instructions file published in the GitHub repository aliildan/cochrane-mcp (1 stars, last pushed 2mo ago), licensed MIT. It adds 2,384 tokens to every session, about $0.0119 per session on Opus 5. A static security scan graded it C with 2 findings (recursive force delete, makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.

Research Scientific

next.js AGENTS.md

AGENTS.md instructions for vercel/next.js, covering next.js development guide, codebase structure, monorepo overview, core package: packages/next and other important packages.

vercel/next.js · 7,296 tokens

codex AGENTS.md

AGENTS.md instructions for openai/codex, covering rust/codex-rs, the codex-core crate, code review rules, crate api surface and model visible context.

openai/codex · 5,153 tokens

vscode buildNext.instructions.md

Working notes and architecture documentation for the new esbuild-based build system in build/next. Use when making changes to the new build pipeline (transpile/bundle commands, NLS plugin, source-map handling, resource copying, or self-hosting watch tasks).

microsoft/vscode · 6,785 tokens

cochrane-mcp: Instructions file for Claude Code