ActivityWatch MCP Server
@8bitgentleman
About ActivityWatch MCP Server
Model Context Protocol server for ActivityWatch time tracking data
Basic information
Config
No standard config provided
This server doesn't expose a parseable MCP config block in its README. See the repository for install instructions.
RepositoryTools
No tools detected
We auto-extract tools from the README. The maintainer can list them under a ## Tools heading to populate this section.
Overview
What is ActivityWatch MCP Server?
A Model Context Protocol server that connects to ActivityWatch, allowing LLMs like Claude to interact with your time tracking data.
How to use ActivityWatch MCP Server?
Install via npm or build from source. Configure the server in your MCP client (e.g., Claude for Desktop) by adding it to the client’s config file. ActivityWatch must be running on http://localhost:5600.
Key features of ActivityWatch MCP Server?
- List all ActivityWatch buckets
- Run AQL queries against time data
- Get raw events from any bucket
- Access ActivityWatch configuration settings
Use cases of ActivityWatch MCP Server?
- Get a summary of which applications you used most today
- View your browsing history by time spent per website
- Check how much time you spent in productivity apps
- Retrieve current ActivityWatch settings or a specific setting
- Debug your ActivityWatch data by fetching raw bucket events
FAQ from ActivityWatch MCP Server
What are the prerequisites to use this server?
ActivityWatch must be installed and running, Node.js v14 or higher is required, and an MCP‑compatible client (e.g., Claude for Desktop) is needed.
Why does the server show connection errors?
If ActivityWatch is not running or is not accessible at http://localhost:5600, the server will report connection errors. Ensure ActivityWatch is started and reachable.
How do I fix query errors when using the MCP server?
Check your AQL syntax, verify bucket IDs are correct, confirm the timeperiods contain data, and review ActivityWatch logs for details. Common formatting issues include splitting query statements into separate array elements instead of one string.
What is the correct format for queries?
All query statements must be in a single string inside the query array. Example: ["events = query_bucket('bucket_id'); RETURN = events;"]. timeperiods should be an array of date‑range strings like ["2024-10-28/2024-10-29"].
How does the server connect to ActivityWatch?
The server connects to the ActivityWatch API at http://localhost:5600 by default. If your ActivityWatch instance uses a different host or port, modify the source code.
More Other MCP servers
ICSS
chokcoco不止于 CSS
Blender
ahujasidOpen-source MCP to use Blender with any LLM
ghidraMCP
LaurieWiredMCP Server for Ghidra

Peekaboo MCP – lightning-fast macOS screenshots for AI agents
steipetePeekaboo is a macOS CLI & optional MCP server that enables AI agents to capture screenshots of applications, or the entire system, with optional visual question answering through local or remote AI models.
ACI: Open-Source Infra to Power Unified MCP Servers
aipotheosis-labsACI.dev is the open source tool-calling platform that hooks up 600+ tools into any agentic IDE or custom AI agent through direct function calling or a unified MCP server. The birthplace of VibeOps.
Comments