Acrawl

Please install the `acrawl` MCP server into my current AI client (that's you).

Required prerequisites (do these first if not already done):
- **Install acrawl binary** — Install the acrawl CLI using the platform install script so the `acrawl` command is available. Run: `curl -fsSL https://raw.githubusercontent.com/Mingye-Lu/AgenticCrawler/main/install.sh | bash` (https://github.com/Mingye-Lu/AgenticCrawler)
- **Install Node.js 20+** — Required for browser features; the installer sets up CloakBrowser for browser automation. (https://nodejs.org/)

Optional prerequisites:
- Configure an LLM provider for run_goal — The MCP server's `run_goal` tool requires acrawl credentials in `~/.acrawl/credentials.json`. Set up a provider interactively, such as Anthropic or OpenAI. Run: `acrawl auth anthropic`

Canonical MCP server config (stdio transport):
- command: `acrawl`
- args: ["mcp"]
- optional environment variables:
  - `ACRAWL_CONFIG_HOME`: Override the directory where acrawl stores configuration and credentials instead of the default `~/.acrawl`. Useful if the MCP client runs in an environment with a different home directory. (example: `<your-acrawl-config-dir>`)

Note: acrawl's built-in MCP server is stdio-only in this release. It exposes direct browser tools without extra configuration. The `run_goal` autonomous tool additionally requires acrawl to have provider credentials configured in `credentials.json`.

Add this MCP server to my current client's config in the correct format for you. If you need secrets or credentials I haven't provided, ASK me — do not invent values or leave raw placeholders. After adding it, tell me how to verify the server is connected.