Skip to directory

Agent Hub MCP

gilbarbara

stdio12+ documented tools0

About Agent Hub MCP

Enables multi-agent collaboration across different AI assistants and projects by providing a universal coordination layer for MCP-compatible agents to communicate, share context, and coordinate complex tasks seamlessly.

What you can use it for

  • developer tools
  • productivity
Connection and setup
Hosting
Local stdio process
Transport
stdio
Endpoint
Local process; no hosted endpoint documented
Authentication
No credentials were configured; the tested read-only tool “get_hub_status” succeeded in an isolated environment. Other tools may have additional requirements.

Requirements

  • Node.js 18+
  • The npm package repository matched the repository linked from the public profile.

Setup steps

  1. npx -y agent-hub-mcp@0.4.0
Tools 12+

Twelve stored tools remain a snapshot, not independently verified in this pass.

  • register_agent

    Register an agent with the hub

    Parameters: id, projectPath, role, capabilities, collaboratesWith
    01
  • send_message

    Send a message to another agent or broadcast to all agents

    Parameters: from, to, type, content, metadata, priority, threadId
    02
  • get_messages

    Retrieve messages for an agent

    Parameters: agent, markAsRead, type, since
    03
  • get_hub_status

    Get overview of hub activity, agents, and collaboration opportunities

    04
  • create_feature

    Create a new feature for multi-agent collaboration

    Parameters: name, title, description, priority, estimatedAgents, createdBy
    05
  • create_task

    Create a task within a feature with agent delegations

    Parameters: featureId, title, description, delegations, createdBy
    06
  • create_subtask

    Create implementation subtasks within a delegation

    Parameters: featureId, delegationId, subtasks, createdBy
    07
  • get_features

    Get list of features with optional filtering

    Parameters: status, priority, agent, createdBy
    08
  • get_feature

    Get complete feature data including tasks, delegations, and subtasks

    Parameters: featureId
    09
  • accept_delegation

    Accept a delegation assigned to an agent

    Parameters: featureId, delegationId, agentId
    10
  • update_subtask

    Update subtask status and provide output/context

    Parameters: featureId, subtaskId, status, output, blockedReason, updatedBy
    11
  • sync

    Comprehensive sync with the hub - get messages, workload, and status in one call

    Parameters: agentId, markAsRead
    12

This server’s available tools can vary by account, permissions, or configuration. The list shown is not its full runtime catalog.

Limitations and notes
  • Runtime verification covered one read-only tool call (get_hub_status); other listed tools were not individually exercised.
Original sources
  • MCP.Directory public profile

    Name, description, publisher identifier, categories, transport, tool inventory, and install configuration.

  • Linked publisher repository

    Repository URL linked from the public profile; npm package repository metadata matched this GitHub repository.

  • npm registry metadata

    Published package version 0.4.0 checked before launch. Registry repository: git+https://github.com/gilbarbara/agent-hub-mcp.git.

  • maintainer repository documentation

    Maintainer repository confirms an MCP server for communication and coordination across AI agents; complete inventory not confirmed in retrieved evidence.

Documentation checked 2026-10-28T00:00:00Z. Connection has not been runtime tested.