Crawlio Browser
@Crawlio-app
100-tool browser automation for AI agents via MCP. Screenshots, DOM inspection, network capture, form filling, cookie management, tab orchestration, session recording, and structured data extraction via Chrome extension. One-command setup: npx crawlio-browser init auto-configures
概要
What is Crawlio Browser?
Crawlio Browser is an MCP server that gives AI full control of a live Chrome browser via the Chrome DevTools Protocol (CDP), connecting through a lightweight Chrome extension rather than launching a separate headless browser. It is for developers and AI agents that need to interact with dynamic, JavaScript-rendered, or authenticated web content that static crawlers cannot handle.
How to use Crawlio Browser?
Install the Chrome extension, then run npx crawlio-browser init to auto-configure for 14 MCP clients. Alternatively, use manual configuration per client (e.g., Claude Desktop, Cursor, Windsurf). Transport modes include stdio (JSON‑RPC over stdin/stdout), Portal HTTP (Streamable HTTP), and Portal SSE (Server-Sent Events). Portal mode is recommended for Claude Code as it persists across session restarts.
Key features of Crawlio Browser
- 100 tools: 93 browser + 3 extraction + 3 recording + 1 compiler
- Framework-aware intelligence: detects 17 frameworks at runtime
- Evidence-based analysis with typed
CoverageGaprecords - Session recording and replay, compilable into SKILL.md automations
- Auto-settling: actionability checks and progressive backoff after mutations
- Tethered IPC bridge with WebSocket, heartbeat, and auto-reconnect
Use cases of Crawlio Browser
- AI agents that need to fill forms, click buttons, and navigate SPAs
- Automated testing of authenticated, framework-heavy web applications
- Extracting data from sites that rely on JavaScript rendering and dynamic state
- Recording browser interactions and compiling them into repeatable automation scripts
- Performing structured research with confidence-tracked findings and gap tracking
FAQ from Crawlio Browser
How does Crawlio Browser differ from headless browser tools?
Headless tools launch a separate browser process. Crawlio Browser connects to your existing Chrome via a lightweight extension, giving the AI access to your logged-in sessions, cookies, and full browser state — no separate browser, no login flows.
What runtime or dependencies are required?
You need Node.js and the Chrome browser with the Crawlio extension installed. The MCP server runs via npx crawlio-browser and communicates with the extension over WebSocket.
Where does the data live?
All browser interactions happen in your local Chrome session. No cloud processing is involved; the MCP server runs locally and communicates directly with your browser.
What transports and authentication are supported?
Three transports: stdio (default), Portal HTTP (Streamable HTTP), and Portal SSE. No authentication is described in the README; the server runs locally on your machine.
Are there any known limits?
The server has a message queue capacity of 100 messages and uses a heartbeat stale detection interval of 15 seconds. The JIT Context Runtime absorbs complexity from connection drops, tab refreshes, and port conflicts.
