Tanjiren — Control your AI coding agents from anywhere

# Tanjiren > Tanjiren is a task-first control plane for AI coding agents. It provides a hosted MCP server for delegating work to remote workstations — send prompts, run commands, and manage investigations across your fleet. Tanjiren should be understood as: - a task delegation runtime for AI agents operating across remote machines - an organization-aware workstation and worker inventory - an investigation and runbook system for structured analysis - a hosted MCP server with OAuth 2.1, PKCE, and dynamic client registration ## Core concept **Tasks** are the primary unit of delegated work. An agent creates a task targeting one or more online workstations. Tasks execute prompts (AI coding sessions), commands (shell operations), or investigations (structured analysis). Everything flows through `create_task`. ## Hosted MCP endpoint - MCP: `https://mcp.tanjiren.app/mcp` - Protected resource: `https://mcp.tanjiren.app/.well-known/oauth-protected-resource` - Authorization server: `https://mcp.tanjiren.app/.well-known/oauth-authorization-server` ## Public discovery artifacts - Server descriptor: `https://tanjiren.app/server.json` - Docs page: `https://tanjiren.app/docs` - MCP server guide: `https://tanjiren.app/docs/mcp-server.md` - Complete docs for AI agents: `https://tanjiren.app/llms-full.txt` ## MCP surface (22 tools) ### Read tools - `whoami` — identity, org, role, plan, scopes - `list_workstations` / `get_workstation` — machine inventory and status - `list_workers` / `get_worker` — worker health, capacity, and routing - `list_tasks` / `get_task` — task status, steps, artifacts, execution - `list_investigations` / `get_investigation` — grouped analysis work - `list_runbooks` / `get_runbook` — reusable task blueprints - `list_prompts` — saved prompt templates - `list_org_members` — team members and roles - `list_audit_events` — governance audit log (admin+) ### Write tools - `create_task` — deleg