PyreCrawl

13 MCP tools for AI agents: scrape, extract, crawl, map, search, academic papers (arXiv + Crossref), batch scrape, deep research, monitor page changes, persistent browser session, cache control. 3-tier auto-fallback ladder with free Cloudflare bypass, no API keys. Self-hosted Firecrawl alternative. `uvx pyrecrawl serve`

GitHub

Try this MCPSponsored

Documentation

🔥 PyreCrawl — Web Browsing Superpowers for Your AI Agent

One command gives any AI agent the whole web. Scrape, extract, crawl, map, and search — self-hosted, no API keys, no rate limits, no subscription.

PyreCrawl speaks MCP (Model Context Protocol), the standard tool interface for Claude, Cursor, VS Code, Codex, OpenCode, Hermes, and any MCP-compatible agent.

A smart auto-fallback ladder always picks the cheapest method that succeeds:

fast HTTP
    │  (403/503/Cloudflare challenge or empty body)
    ▼
stealth browser (real Chromium + Cloudflare solver)
    │  (still blocked, or the page needs full JS rendering)
    ▼
deep processing (LLM-ready markdown, citations, structured extraction)

⚡ Tools exposed

Tool	What it does
`scrape(url, prefer="auto")`	Single URL → LLM-ready markdown
`extract(url, schema)`	Scrape + structured extraction (JsonCss schema)
`map_site(root, include_pattern=None, limit=200)`	Enumerate all internal URLs
`crawl(root, max_pages=5, prefer="auto", include_paths=None, exclude_paths=None, max_depth=0)`	Multi-page crawl with path filters + true BFS depth
`document(url)`	PDF/DOCX/PPTX → markdown (no browser, optional `[docs]` extras)
`search(query, limit=10)`	Web search via DuckDuckGo HTML (no API key)
`search_papers(query, limit=8, source="arxiv", category=None)`	Academic search via arXiv + Crossref (no API key) — feed `pdf_url` into `document`
`batch_scrape(urls[], ...)`	Many URLs in ONE call — parallel, deduped, cache-aware
`deep_research(query, limit=5, scrape_top=3)`	Search → evidence pack with [n] citations (no LLM synthesis — your agent does that)
`monitor(url, action, css_selector=None)`	Change detection with persisted snapshots + unified diff
`session(session, action, ...)`	Persistent browser session (cookies kept) — login walls, multi-step flows, screenshots
`cache(action)`	Inspect/clear/enable/disable the HTTP response cache
`health()`	Versions + import sanity check

MCP Resources (read-only state without a tool call): pyrecrawl://cache/stats · pyrecrawl://sessions · pyrecrawl://monitors

MCP Prompts (ready-made playbooks): research(topic) · rag_ingest(site) · watch_page(url)

Env flags

Variable	Default	Effect
`PYRECRAWL_CACHE`	off	`1` = in-memory LRU (128 pages), or a directory path (reserved for disk mode)
`PYRECRAWL_CACHE_TTL`	`900`	Cache entry lifetime in seconds
`PYRECRAWL_MONITOR_DIR`	`~/.pyrecrawl/monitors`	Where monitor snapshots persist

prefer options: "auto" (default ladder) · "fast" (HTTP only) · "stealth" (CF bypass) · "llm" (deep processing).

🚀 Install & Use (one-liner)

1. Install

UV (recommended — one command, zero Python setup)

UV is a fast Python package manager that handles Python itself — no need to install Python separately. Get it once:

# macOS / Linux
curl -LsSf https://astral.sh/uv/install.sh | sh

# Windows (PowerShell)
powershell -ExecutionPolicy ByPass -c "irm https://astral.sh/uv/install.ps1 | iex"

Learn more about UV →

Then run PyreCrawl directly — no venv, no pip install, no Python download:

uvx pyrecrawl@latest

Or via uv tool install (persistent, recommended for regular use)

uv tool install pyrecrawl

Or via pipx (alternative)

pipx install pyrecrawl

Or via pip into a venv

pip install pyrecrawl

2. One-time browser engines

pyrecrawl setup

This installs Chromium + stealth browser engines (~2 min, one-time).

3. Register with your AI agent

# Auto-detect installed agents and write their MCP configs
pyrecrawl install

# Or target specific agents
pyrecrawl install claude-desktop cursor

# Dry-run to preview what would change
pyrecrawl install --dry-run

Supported agents: claude-desktop, claude-code, cursor, vscode, codex, opencode, hermes.

4. Start chatting

After installing + registering, restart your agent (or start a new session). Then ask:

"Scrape https://example.com and summarize it."

The tools appear as mcp_pyrecrawl_scrape, mcp_pyrecrawl_extract, mcp_pyrecrawl_map_site, mcp_pyrecrawl_crawl, mcp_pyrecrawl_search, mcp_pyrecrawl_health.

📚 Manual config (if `pyrecrawl install` doesn't match your setup)

Claude Desktop

Config file

Linux: ~/.config/Claude/claude_desktop_config.json
macOS: ~/Library/Application Support/Claude/claude_desktop_config.json
Windows: %AppData%\Claude\claude_desktop_config.json

{
  "mcpServers": {
    "pyrecrawl": {
      "command": "uvx",
      "args": ["--from", "pyrecrawl", "pyrecrawl", "serve"]
    }
  }
}

Claude Code

Config file: project-scoped .mcp.json

{
  "mcpServers": {
    "pyrecrawl": {
      "command": "uvx",
      "args": ["--from", "pyrecrawl", "pyrecrawl", "serve"]
    }
  }
}

Cursor

Config file: ~/.cursor/mcp.json

{
  "mcpServers": {
    "pyrecrawl": {
      "command": "uvx",
      "args": ["--from", "pyrecrawl", "pyrecrawl", "serve"]
    }
  }
}

VS Code / Copilot

Config file: .vscode/mcp.json (project-scoped)

{
  "servers": {
    "pyrecrawl": {
      "command": "uvx",
      "args": ["--from", "pyrecrawl", "pyrecrawl", "serve"],
      "type": "stdio"
    }
  }
}

Codex CLI

Config file: ~/.codex/config.toml

[mcp_servers.pyrecrawl]
command = "uvx"
args = ["--from", "pyrecrawl", "pyrecrawl", "serve"]

OpenCode

Config file: ~/.config/opencode/opencode.json

{
  "mcp": {
    "pyrecrawl": {
      "type": "local",
      "command": ["uvx", "--from", "pyrecrawl", "pyrecrawl", "serve"],
      "enabled": true
    }
  }
}

Hermes

Config file

Linux/macOS: ~/.hermes/config.yaml
Windows: %LocalAppData%\hermes\config.yaml

mcp_servers:
  pyrecrawl:
    command: uvx
    args:
      - --from
      - pyrecrawl
      - pyrecrawl
      - serve
    enabled: true

Windows note: uvx must be on PATH. If not, use the full path to uvx.exe (e.g. C:\Users\<you>\AppData\Local\hermes\bin\uvx.exe).

🧠 How the ladder chooses

PyreCrawl runs each request through three tiers, stopping at the first one that returns a complete, LLM-ready result:

Concern	Fast tier	Stealth tier	Deep tier
Static HTML page	✅ ~200ms	—	—
Cloudflare-protected	❌	✅ Turnstile solver	—
JS-heavy SPA	❌	✅ real Chromium	—
Live DOM data (input `.value`, JS state)	❌	✅ `js` param	—
LLM-ready markdown + citations	—	—	✅ BM25, fit-markdown
Structured extraction (CSS schema)	—	—	✅
Deep crawl (BFS/DFS/BestFirst)	—	—	✅ adaptive

The agent never has to pick. prefer="auto" does it every call.

Live DOM data with `js` and `wait_for`

Some sites keep the data you want in a DOM property (e.g. an <input>'s .value) that JS writes after an XHR — it never appears in the serialized HTML. The scrape tool accepts two stealth-tier params for exactly this:

{
  "url": "https://temp-mail.org/id",
  "prefer": "stealth",
  "wait_for": "document.getElementById('mail').value.includes('@')",
  "js": "document.getElementById('mail').value"
}

wait_for — a JS predicate expression polled until truthy (bounded by timeout). Use it instead of guessing a sleep for anything that arrives asynchronously.
js — a JS expression evaluated once the page settles; the value comes back in meta.js_result. Errors are captured in meta.js_error (the page result is still returned, never a crash).

📊 Compared to Firecrawl (hosted)

	Firecrawl	PyreCrawl
Cost	Free 1k/mo, then $16–333/mo	Free, self-hosted
Local LLM support	❌	✅ Ollama / any LLM
Cloudflare bypass	✅ (Fire-Engine, paid)	✅ (free, built-in)
Markdown + BM25	✅	✅
Self-host	❌	✅
Academic paper search	❌	✅ arXiv + Crossref (`search_papers`)
Hosted search API	✅ /search	⚠️ DuckDuckGo HTML + arXiv/Crossref (no key)

🔧 Development

git clone https://github.com/SanggonBoy/PyreCrawl.git
cd PyreCrawl
uv venv --python 3.12 .venv
source .venv/Scripts/activate  # Windows; or .venv/bin/activate on macOS/Linux
uv pip install -e ".[dev]"
python -m playwright install chromium
scrapling install

Run tests

python scripts/selfcheck.py   # real-network smoke test
python scripts/probe_stdio.py # stdio JSON-RPC probe

📦 Publish

Maintainers only:

git tag v0.8.0
git push origin v0.8.0

GitHub Actions builds + uploads to PyPI via trusted publishing.

🔔 Stay up to date

PyreCrawl checks PyPI on every startup and reports the latest version — your MCP agent sees this automatically via the health() tool response and can notify you inline.

To check manually:

pyrecrawl version

To upgrade:

pyrecrawl update   # runs: uv tool upgrade pyrecrawl

Get notified of new releases: click Watch → Releases only at the GitHub repo to receive email notifications when a new version is published.

📜 Uninstall

# Remove from all agent configs
pyrecrawl uninstall

# Remove the package
uv tool uninstall pyrecrawl

🛡️ License

MIT — see LICENSE.

🙏 Credits

Built on the shoulders of Scrapling and Crawl4AI — both MIT, both excellent.