The Risal MCP server lets AI agents (Claude and any MCP-compatible client) work with your tasks, sprints, timers and servers through the Model Context Protocol.
1. Get an API key
In the Risal app, create a personal access token (PAT). Scope it to only the capabilities you want the agent to have, then copy the risal_pat_… value.
2. Add the server
npx (nothing to install)
claude mcp add risal --scope user \
--env RISAL_API_KEY=risal_pat_… \
-- npx -y risal-mcp-server
Docker
claude mcp add risal --scope user \
--env RISAL_API_KEY=risal_pat_… \
-- docker run -i --rm -e RISAL_API_KEY \
shaxzod/risal-mcp-server:latest
To point at a self-hosted instance, set RISAL_API_BASE to your API host.
3. Use it
Ask your agent in natural language, for example:
“Create task NOT-42, start a sprint, and log one hour against it.”
The agent calls the matching MCP tools directly.
What the agent can do
- Create and update tasks (single or bulk), and parse free text into tasks
- Manage projects and sprints
- Start/stop timers and log time
- Read AI-usage statistics
- Operate servers: list hosts and metrics, read Docker/PM2 logs, restart services
Security
Everything runs through your token — the agent can never exceed its scope. Rotate the token whenever you like; if you do, update it everywhere the server is configured.
The package is published on npm and Docker Hub.