Skip to main content
Glama
cerase-ai

Cerase Tasks MCP

Official
by cerase-ai
OverviewSchemaSourceRelated ServersDiscussionsAdmin
Python
Remote

cerase-tasks MCP

The agent's Projects & Tasks board (PROJ-3). A thin first-party proxy to the control-plane PROJ-2 internal endpoints — it owns no state itself; the durable, Filament-renderable board lives in the control-plane Postgres (Decision B). Same delivery pattern as cerase-memory / cerase-transcriber: the policy gateway routes to it and injects the calling Agent's agent_id (un-spoofable), so the agent can only ever touch its own board.

Tools

  • create_task(agent_id, title, project?, phase?) → task — record a unit of work the user would recognise (NOT a micro-step). Lands in the default "General" project unless project (a project id) is given.

  • set_status(agent_id, task_id, status) → task — move along backlog/todo/doing/review/done; the board stamps elapsed time.

  • list_tasks(agent_id, project?) → {tasks} — compact open list (id, title, status, phase); call on demand to re-orient.

  • create_project(agent_id, name) → project.

agent_id is required and gateway-bound — the server raises on empty.

Related MCP server: Agent Kanban

Boundaries

mem0 = facts the agent knows · workspace = the actual files · this board = the STATE of the work. Micro-steps (how the agent does it) stay in OpenCode's ephemeral native todo, never here.

Env

  • CERASE_CONTROL_PLANE_URL (default http://cerase-control-plane:8000)

  • CERASE_INTERNAL_SECRET (the internal-bearer shared secret)

GHCR image ghcr.io/cerase-ai/cerase-tasks-mcp:<tag>.

This server cannot be deployed

One endpoint for every MCP toolRayrun runs every MCP server behind one URL and holds the credentials. Each call is decided per client, tool, and argument, then recorded.ray.runAd

Maintenance

ActivityMaintained
ResponsivenessNo issues

Resources

Related MCP Connectors

Related MCP Servers

  • pith

    SiluPanda
    A
    license
    Not graded
    quality
    D
    maintenance
    MCP server for task management that enables AI agents to read, create, update tasks, and track work sessions, allowing agents and humans to collaborate on the same task board.
    6 npm
    9
    MIT

  • Agent Kanban

    graywrk
    A
    license
    Not graded
    quality
    B
    maintenance
    An AI-native kanban board MCP server where agents pull tasks via the Model Context Protocol. It provides tools for task discovery, claiming, progress updates, and review workflows.
    4
    MIT

  • agents-board

    fmilioni
    A
    license
    Not graded
    quality
    B
    maintenance
    MCP server that provides a real-time project management board with cards, sprints, comments, and docs, enabling coding agents to query and edit project state.
    MIT