This MCP server acts as a bridge between Claude and a project management backend, exposing a comprehensive set of tools to manage projects, tasks, users, teams, organizations, tags, comments, and files.
Health Check
Check API liveness with pm_health.
Projects
Create, list, get, update, and delete projects.
Upload, download, list, update, and delete files attached to projects (base64-encoded transfers).
Tasks
Create, list, get, update, and delete tasks.
Filter by status, priority, project, owner, tags, overdue, starred, and more.
Attach and detach tags to/from tasks.
Comments
Create, list, update, and delete comments on tasks.
Users
Create, list, get, update, and delete users (some actions admin-only).
Attach and detach contact labels to/from users (admin-only).
Export users as joined plain text or CSV with configurable fields and filters.
User Labels (Contact Labels)
Create, list, get, and delete global contact labels.
Organizations
Create, list, and get organizations.
Teams
Create, list, and get teams, optionally filtered by organization.
Tags
Create, list, get, update, and delete tags, optionally scoped to a project or global.
Click on "Deploy Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type @ followed by the MCP server name and your instructions, e.g., "@project-manager Create a task 'Fix login bug' in project 'MainApp'"
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
This project implements stdio MCP server, so that Claude can search data by sending requests to an API backend running on localhost:8000
Documentation: docs/ — implementation overview and full pm_* tool reference.
This is the entry point for your MCP server. Breaking it down:
@click.command() + @click.option — CLI interface with two flags:
--base-url: override the API URL without changing env vars (e.g. --base-url http://staging:8000)
--transport: how Claude communicates with the server — stdio (default, used by Claude Desktop via stdin/stdout pipes), sse or streamable-http for network-based transports
def main(...):
Creates a ProjectManagerClient — your HTTP client that talks to the FastAPI backend
Calls build_mcp(client) — registers all the pm_* tools with the MCP framework, wrapping the client's methods as Claude-callable tools
mcp.run(transport=transport) — starts the server loop, listening for tool calls from Claude
client.close() in finally — cleanly shuts down the HTTP client when the process exits
Why stdio is the default: Claude Desktop launches this as a subprocess and communicates via stdin/stdout pipes — no network port needed. The sse/streamable-http options would be for remote deployments where Claude connects over HTTP instead.
In short: this boots the bridge between Claude and your project manager API, using whichever communication channel fits the deployment context.
You add below into mcpServers section of %APPDATA%\Claude\claude_desktop_config.json
What does this achieve in %APPDATA%\Claude\claude_desktop_config.json
"project-manager": {
"command": "uv",
"args": [
"run",
"--directory",
"c:\\Users\\amrusub\\programming\\pet_projects\\project_manager_mcp",
"project-manager-mcp"
],
"env": {
"PROJECT_MANAGER_API_BASE": "http://127.0.0.1:8000"
}
}
},This registers a local MCP server called "project-manager" with Claude Desktop. Here's what each part does:
command: "uv" — uses uv (the fast Python package manager) to run the server
args — tells uv to run the project-manager-mcp entry point from your local project directory at c:\Users\amrusub\programming\pet_projects\project_manager_mcp
env — sets an environment variable so the MCP server knows to talk to your locally running API at http://127.0.0.1:8000
Net effect: When Claude Desktop starts, it spins up your local project-manager-mcp Python process in the background. Claude then gets access to whatever tools that server exposes (in your case, the project-manager:pm_* tools visible in this conversation — create tasks, list projects, manage users, etc.), all backed by your local FastAPI server on port 8000.
So it's essentially a bridge: Claude ↔ MCP server process ↔ your local REST API.
POST /comments — TaskCommentCreate.
Inputs: payload (required), x_user_id. Returns: result
POST /organizations — OrganizationCreate {name}.
Inputs: payload (required), x_user_id. Returns: result
POST /projects — body matches ProjectCreate (name, description_md, owner, optional slug, color_key, sort_order, settings_json, owner_user_id).
Inputs: payload (required), x_user_id. Returns: result
POST /tags — TagCreate (name, optional project_id).
Inputs: payload (required), x_user_id. Returns: result
POST /tasks — TaskCreate (project_id, title, short_name, plus optional fields).
Inputs: payload (required), x_user_id. Returns: result
POST /teams — TeamCreate {name, optional organization_id}.
Inputs: payload (required), x_user_id. Returns: result
POST /users — admin only. UserCreate: display_name; optional email, phone, avatar_url, presence, role (admin|member), user_kind (internal|client), organization_id, team_id.
Inputs: payload (required), x_user_id. Returns: result
POST /user-labels — UserLabelCreate {name}.
Inputs: payload (required), x_user_id. Returns: result
DELETE /comments/{comment_id}.
Inputs: comment_id (required), x_user_id. Returns: result
DELETE /projects/{project_id}.
Inputs: project_id (required), x_user_id. Returns: result
DELETE /projects/{project_id}/files/{file_id}.
Inputs: file_id (required), project_id (required), x_user_id. Returns: result
DELETE /tags/{tag_id}.
Inputs: tag_id (required), x_user_id. Returns: result
DELETE /tasks/{task_id}.
Inputs: task_id (required), x_user_id. Returns: result
DELETE /users/{user_id} — admin only.
Inputs: user_id (required), x_user_id. Returns: result
DELETE /user-labels/{label_id}.
Inputs: label_id (required), x_user_id. Returns: result
GET /projects/{project_id}/files/{file_id}/download — returns JSON with content_base64, content_type, size_bytes.
Inputs: file_id (required), project_id (required), x_user_id. Returns: result
GET /users/export.csv — CSV export; fields is comma-separated column names; same filters as pm_list_users.
Inputs: fields, team_id, label_id, label_ids, +3 more. Returns: result
GET /users/export/joined — plain text; field: display_name|email|phone; separator: comma|semicolon; same filters as pm_list_users.
Inputs: field, team_id, label_id, label_ids, +4 more. Returns: result
GET /organizations/{organization_id}.
Inputs: organization_id (required), x_user_id. Returns: result
GET /projects/{project_id}.
Inputs: project_id (required), x_user_id. Returns: result
GET /tags/{tag_id}.
Inputs: tag_id (required), x_user_id. Returns: result
GET /tasks/{task_id}.
Inputs: task_id (required), x_user_id. Returns: result
GET /teams/{team_id}.
Inputs: team_id (required), x_user_id. Returns: result
GET /users/{user_id}.
Inputs: user_id (required), x_user_id. Returns: result
GET /user-labels/{label_id}.
Inputs: label_id (required), x_user_id. Returns: result
GET /health — API liveness.
Inputs: no named parameters. Returns: result
GET /comments?task_id=...
Inputs: task_id (required), x_user_id. Returns: result
GET /organizations.
Inputs: x_user_id. Returns: result
GET /projects/{project_id}/files.
Inputs: project_id (required), x_user_id. Returns: result
GET /projects — list projects (optional include_archived).
Inputs: x_user_id, include_archived. Returns: result
GET /tags — optional project_id or global_only.
Inputs: x_user_id, project_id, global_only. Returns: result
GET /tasks — filters optional. task_status: todo|in_progress|blocked|done|cancelled; priority: low|medium|high|urgent.
Inputs: tag_id, priority, x_user_id, project_id, +5 more. Returns: result
GET /teams — optional organization_id filter.
Inputs: x_user_id, organization_id. Returns: result
GET /user-labels — list global contact labels (not task tags).
Inputs: x_user_id. Returns: result
GET /users — optional filters. user_kind: internal|client; label_ids: comma-separated UUIDs (OR); do not combine label_id and label_ids.
Inputs: team_id, label_id, label_ids, user_kind, +2 more. Returns: result
POST /tasks/{task_id}/tags — body {tag_id: ...}; tag_id also accepted as top-level arg for convenience.
Inputs: tag_id (required), task_id (required), x_user_id. Returns: result
DELETE /tasks/{task_id}/tags/{tag_id}.
Inputs: tag_id (required), task_id (required), x_user_id. Returns: result
PATCH /comments/{comment_id}.
Inputs: payload (required), comment_id (required), x_user_id. Returns: result
PATCH /projects/{project_id} — partial ProjectUpdate fields.
Inputs: payload (required), project_id (required), x_user_id. Returns: result
PATCH /projects/{project_id}/files/{file_id} — partial ProjectFileUpdate (name, description_md, kind).
Inputs: file_id (required), payload (required), project_id (required), x_user_id. Returns: result
PATCH /tags/{tag_id}.
Inputs: tag_id (required), payload (required), x_user_id. Returns: result
PATCH /tasks/{task_id} — partial TaskUpdate.
Inputs: payload (required), task_id (required), x_user_id. Returns: result
PATCH /users/{user_id} — partial UserUpdate; requires X-User-Id (self or admin). Fields: display_name, email, phone, avatar_url, presence, role, user_kind, organization_id, team_id.
Inputs: payload (required), user_id (required), x_user_id. Returns: result
POST /projects/{project_id}/files — multipart upload; file bytes as standard base64.
Inputs: filename (required), project_id (required), file_content_base64 (required), kind, +3 more. Returns: result
POST /users/{user_id}/labels — admin only; attach a contact label to a user.
Inputs: user_id (required), label_id (required), x_user_id. Returns: result
DELETE /users/{user_id}/labels/{label_id} — admin only.
Inputs: user_id (required), label_id (required), x_user_id. Returns: result
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
Loading tool schema changelog…
Scored across 46 tools
Each tool targets a distinct resource and action (e.g., pm_create_comment vs pm_create_task). The names and descriptions make it clear which entity is being operated on, with no overlapping functionality. The distinction between pm_list_tags and pm_list_user_labels is explicit.
All tools follow the consistent pattern `pm_verb_noun`. Verbs are standard (create, get, list, update, delete, upload, download, export, attach, detach) and nouns correspond to resources. No mixed conventions or vague names.
46 tools is well above the typical well-scoped range (3-15). Even for a comprehensive project manager, this number is high and may overwhelm agents. It borders on excessive, as per calibration (25+ is too many).
The server covers CRUD for many entities (projects, tasks, comments, tags, users, files) but lacks update operations for organizations, teams, and user labels. Also team and organization deletion are missing. These are notable gaps that agents may encounter.
Create projects, nodes, and tasks in UluP Spaces by conversation with Claude.
Manage Redbooth tasks, task lists, comments, tags and files from Claude and other AI assistants.
Lets Claude, ChatGPT, and Muse read and update your beckett tasks, notes, calendar, and lists.
Read your tracked time and projects, pause tracking or add a note, from Claude or ChatGPT.
