with one click
mcp-adapter
Use the bundled pi-mcp-adapter to connect MCP servers through .pi/mcp.json.
Menu
Use the bundled pi-mcp-adapter to connect MCP servers through .pi/mcp.json.
Generate a 7-day token usage chart (all chats) and post it to the web UI timeline.
Search the web via SearXNG (or DuckDuckGo fallback) and optionally fetch result pages as raw Markdown content.
Search via SearXNG (or DuckDuckGo fallback), fetch top results, and return sentence-level summaries plus optional converted page content.
Diagnose and fix issues with the Pibox container environment — tools, paths, mounts, pi configuration.
Reinstall piclaw from workspace source and force-restart the running process. Use after making code changes to piclaw.
Situate yourself by generating a 1-page situation report and maintaining Obsidian-style daily summary notes.
| name | mcp-adapter |
| description | Use the bundled pi-mcp-adapter to connect MCP servers through .pi/mcp.json. |
| distribution | public |
PiClaw ships pi-mcp-adapter, which exposes MCP servers to the agent through a single mcp tool plus /mcp commands.
Preferred project-local config:
/workspace/.pi/mcp.json
Starter example:
cp /workspace/.pi/mcp.json.example /workspace/.pi/mcp.json
The adapter also understands the global Pi config at:
/config/.pi/agent/mcp.json
Use the project-local file when you want MCP servers tied to the current workspace.
{
"mcpServers": {
"filesystem": {
"command": "bunx",
"args": ["-y", "@modelcontextprotocol/server-filesystem", "."],
"lifecycle": "lazy"
}
}
}
mcp({})
mcp({ search: "read file" })
mcp({ describe: "filesystem_read_file" })
mcp({ tool: "filesystem_read_file", args: "{\"path\":\"./README.md\"}" })
args must be a JSON string.
/mcp
/mcp tools
/mcp reconnect
/mcp reconnect filesystem
/mcp-auth <server>
.pi/mcp.json, restart the session or reload Pi so the adapter re-reads config.directTools only for small, high-value tool sets. Large MCP servers are usually better left behind the mcp proxy.