mcpField
P
ProjectBrain MCP

ProjectBrain MCP

Remote
github.com/9-TrinketsProductivity0Updated 28w ago
Repository

About ProjectBrain MCP

ProjectBrain MCP is an MCP server for project planning and execution, intended for teams managing project information through ProjectBrain. It exposes tools for tasks, decisions, facts, milestones, comments, and team messages over Streamable HTTP, with bearer-token authentication for tool use and data access.

Key features

  1. 01Task and milestone management
  2. 02Decision and fact tools
  3. 03Comments and team messaging
  4. 04Streamable HTTP endpoint
  5. 05Bearer-token authentication

Use cases

  1. 01Managing project tasks and milestones
  2. 02Recording project decisions and facts
  3. 03Sharing comments and team messages

MCP support

Exposes ProjectBrain tools to MCP clients through a Streamable HTTP endpoint. Discovery methods are available without authentication; tool execution and data access require a bearer token.

#project-management#tasks#decisions#knowledge-base#team-collaboration#mcp

Configuration

Remote servers need no local install. Pick your client, then copy the config or install in one click.

Endpointhttps://mcp.projectbrain.tools
Copy install command
Claude CodeTerminal
Cursor.cursor/mcp.json
VS Code.vscode/mcp.json
Claude Desktopclaude_desktop_config.json
Windsurf~/.codeium/windsurf/mcp_config.json

Run commands in a terminal; paste JSON into the client’s MCP config and restart. Servers that need sign-in will prompt on first connect.

All endpoints
Streamable HTTPhttps://mcp.projectbrain.toolsAuth required

Tools

6
ParameterTypeRequiredDescription
actionstringNoContext action: session, summary, changes, search, or shortlist.
project_idanyNoProject UUID used by session/summary/changes/search actions.
sinceanyNoISO-8601 timestamp used by changes action to bound results.
qanyNoSearch query used by search/shortlist actions.
limitintegerNoMaximum number of results to return for list-like context actions.
full_tool_modebooleanNoWhen true, shortlist includes full operation catalog instead of top-ranked subset.
task_idanyNoOptional task UUID for context-sensitive knowledge retrieval.
intentanyNoOptional user intent for semantic knowledge retrieval.
knowledge_limitintegerNoMax number of knowledge items to return when intent is provided.
response_modestringNoOutput format: human, json, or both (where supported).

Details

From the project README

ProjectBrain MCP

ProjectBrain MCP is a Model Context Protocol (MCP) server for project planning and execution workflows. It exposes tools for tasks, decisions, facts, milestones, comments, and team messaging through a Streamable HTTP endpoint.

What it does

  • Serves MCP tools over HTTPS at https://mcp.projectbrain.tools
  • Allows unauthenticated MCP discovery requests for initialize, notifications/initialized, ping, tools/list, resources/list, resources/templates/list, and prompts/list
  • Authenticates bearer tokens (JWT or API key) for all tool execution and data access
  • Provides MCP OAuth metadata endpoints
  • Executes tool actions against the ProjectBrain API

Service endpoints

  • GET /health
  • GET /.well-known/oauth-protected-resource
  • GET /.well-known/oauth-authorization-server
  • POST /

Configuration

Set these environment variables:

  • SERVER_URL (default: http://localhost:8000)
    Base URL of the ProjectBrain API.
  • MCP_SERVER_URL (default: same as SERVER_URL)
    Public base URL used in OAuth resource metadata.
  • JWT_SECRET_KEY
    Secret used to validate JWT bearer tokens.
  • JWT_ALGORITHM (default: HS256)
  • CORS_ORIGINS (default: ["http://localhost:5173"])
    Accepts either a JSON array or a comma-separated list.
  • MCP_REQUEST_TIMEOUT_SECONDS (default: 30)
  • MCP_UNAUTH_DISCOVERY_METHOD_SCAN_BYTES (default: 8192)
    Max unauthenticated request bytes scanned to detect JSON-RPC method in lightweight middleware parsing.
  • SENTRY_DSN
    Sentry DSN for MCP service error/performance reporting.
  • SENTRY_TRACES_SAMPLE_RATE (default: 0.2)
    Trace sample rate passed to sentry_sdk.init.
  • SENTRY_SEND_DEFAULT_PII (default: false)
    Set to true to include richer MCP request/response context with MCPIntegration (review data collection implications before enabling).
  • SENTRY_ENVIRONMENT (optional)
    Sentry environment tag (for example: production, staging).

Local development

From repository root:

  1. Install backend dependencies:
    • cd api && uv sync --locked
  2. Start the MCP server:
    • PYTHONPATH=mcp api/.venv/bin/uvicorn api_adapter:app --app-dir mcp --host 0.0.0.0 --port 8001

Directory structure

  • api_adapter.py — FastAPI app entrypoint and auth middleware
  • server.py — MCP tool definitions and HTTP client adapter
  • runtime.py — runtime config and request-scoped auth context

Comments

Markdown supported

New to MCP?

From what the protocol is to installing your first server, these answers cover most first questions.

An open protocol that defines how AI applications connect to external tools and data in a uniform way. Clients talk to the model; MCP servers provide tools, resources and prompt templates. They exchange standard messages, so one server can be reused by many clients.