About GEN
GEN’s hosted MCP server lets MCP-compatible AI systems operate the GEN platform, from agent setup and video generation through editing and publishing. The downloadable PyPI package is deprecated; the hosted endpoint is the maintained service.
Key features
- 01Five-step agent-to-publishing workflow
- 02Tools for ideas and video generation
- 03Video editing and export tools
- 04Social publishing and scheduling
- 05Embedded API reference resource
Use cases
- 01Set up a GEN agent and develop video ideas
- 02Generate, edit, and publish videos
- 03Use GEN from an MCP-compatible AI client
MCP support
The hosted server at https://mcp.gen.pro exposes GEN platform tools to MCP clients over HTTP, authenticated with a Bearer token. The legacy downloadable package is deprecated.
Configuration
Remote servers need no local install. Pick your client, then copy the config or install in one click.
https://mcp.gen.proclaude mcp add --transport http poweredbygen-gen https://mcp.gen.pro{"mcpServers":{"poweredbygen-gen":{"type":"http","url":"https://mcp.gen.pro"}}}Open in Cursor{"servers":{"poweredbygen-gen":{"type":"http","url":"https://mcp.gen.pro"}}}Open in VS Code{"mcpServers":{"poweredbygen-gen":{"type":"http","url":"https://mcp.gen.pro"}}}{"mcpServers":{"poweredbygen-gen":{"serverUrl":"https://mcp.gen.pro"}}}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.
https://mcp.gen.proAuth requiredTools
19| Parameter | Type | Required | Description |
|---|---|---|---|
| actions | array | Yes | One or more typed agent actions: set_profile (brand_name, description, competitors, identity_type, goal, target_platforms, shortform, longform, look, personality, persona_inspirations, research_topics, content_themes, monitored) | set_name (name) | set_location (catalog_version, country_code, city_id; sent alone) | set_primary_voice (default_user_voice {voice_id, source}) | set_primary_photo (the agent profile photo: profile_photo_url) | add_reference_images (agent reference images: content_resource_ids and/or images) | remove_reference_images (ids) | add_linked_account (url, optional display_name) | remove_linked_account (linked_account_id) | complete_profile (description and/or linked_accounts) | create_agent (name, optional description/organization_id) | update_agent (optional name/description) | delete_agent | create_agent_avatar (adds an agent profile photo) | delete_agent_avatar (avatar_id of an agent profile photo) | add_agent_inspiration (url, optional platform) | create_look | connect_agent_elevenlabs (api_key) | reset_agent_profile | share_agent (username_or_email, role viewer|editor|manager) | set_agent_share_role (username_or_email, role viewer|editor|manager) | revoke_agent_share (shared_agent_id) | transfer_agent_ownership (username_or_email) | set_project_share (project_id_or_uuid, permission public|private) |
| agent_id | any | No | The agent ID that owns the assistant; merged into each action exactly where the legacy tool accepts it |
| confirm_token | any | No | Single-use token from a CONFIRMATION_REQUIRED error; repeat the same call with it to complete the destructive action it was issued for. Omit on the first call. |
Details
From the project READMEGEN MCP Server
⚠️ The downloadable package is deprecated — use
https://mcp.gen.proThis is the FINAL PyPI release of
gen-mcp-server. GEN no longer ships a downloadable MCP: the official GEN MCP is the hosted server athttps://mcp.gen.pro. No further versions will be published to PyPI.The package still works so existing integrations have a migration window, but it will not receive new tools, fixes, or API updates — the hosted server is already ahead of it and is the only surface kept in sync with the GEN API. Installing it emits a
DeprecationWarning; the stdio CLI prints a notice to stderr on start.
Use the official hosted MCP (all clients)
No install. Point your MCP client at the hosted endpoint and authenticate with your GEN Personal Access Token (from https://gen.pro — log in, pick an agent, open the API page in the sidebar, click Create API Key):
{
"name": "gen",
"url": "https://mcp.gen.pro",
"headers": { "Authorization": "Bearer gen_pat_…" }
}
(Existing configs using https://mcp.gen.pro/mcp keep working.)
Claude Code:
claude mcp add --transport http gen https://mcp.gen.pro \
--header "Authorization: Bearer gen_pat_…"
Manus: Settings → Integrations → Custom MCP Servers → Add Server
- Server URL:
https://mcp.gen.pro - Authentication: Bearer token → your
gen_…PAT
claude.ai connectors / ChatGPT: add a custom MCP connector with the same URL + bearer token.
The hosted server is multi-tenant — each caller acts as their own GEN identity; your PAT scopes everything you can see and do.
What it is
MCP (Model Context Protocol) server for the GEN platform — lets Claude and any MCP-compatible AI drive the GEN Auto Content Engine + Agent Core: the full 5-step journey from onboarding an agent to publishing a video.
Step 1 Step 2 Step 3 Step 4 Step 5
────── ────── ────── ────── ──────
Set Up → Generate → Idea to → Edit & → Export &
Agent Ideas Vidsheet Generate Publish
Read the gen://api-reference MCP resource for the full teaching document, and
api.gen.pro for the API reference. Tool descriptions
start with the step name (e.g. "Step 4 (Edit & Generate): …") so AI tooling can
route quickly.
Legacy local install (deprecated — migrate to the hosted URL above)
The stdio mode still functions for the migration window:
uvx gen-mcp-server # or: pip install gen-mcp-server && gen-mcp-server
export GEN_API_KEY=your-api-key
It prints a deprecation notice to stderr on start. Do not build new
integrations on it; use https://mcp.gen.pro.
Optional base-URL overrides (legacy stdio only):
export GEN_API_BASE_URL=https://api.gen.pro/v1
export GEN_AGENT_API_URL=https://agent.gen.pro/v1
export GEN_AGENT_CORE_API_URL=https://agent-core.gen.pro/v1
Development (this repo stays alive — it IS the hosted server)
This repository remains the source of the hosted mcp.gen.pro deployment; only
the PyPI distribution is retired.
git clone https://github.com/poweredbyGEN/gen-mcp-server.git
cd gen-mcp-server
uv venv && . .venv/bin/activate
uv pip install -e .
GEN_MCP_PORT=8080 gen-mcp-server --http # hosted streamable-http mode, serves `/` and `/mcp`
See also
- API Docs: api.gen.pro — full API reference and guides
- TypeScript SDK: github.com/poweredbyGEN/gen-typescript-sdk
Changelog
1.0.0 — FINAL PyPI RELEASE (deprecation pointer)
- The downloadable package is deprecated; the official GEN MCP is the hosted
server at
https://mcp.gen.pro(GEN-4785) DeprecationWarningon import when installed from PyPI; stderr notice on stdio CLI start. Hosted (--http) mode is unaffected- Includes everything on main at cut time, including the Python/FastMCP rewrite (GEN-3788) and the 2026-07-30 tool additions (billing, social, schedule, assets, generation, compose — 31 tools) never previously published
0.5.0
- Restructured all 81 tools around the 5-step user journey (GEN-2879)
- Rewrote the embedded
gen://api-referenceresource as a step-by-step teaching document with chained curl examples - Tool descriptions now start with the phase (e.g. "Step 1 (Agent Setup): …") so AI tooling can route quickly
0.4.x
- Added Agent Core flat endpoints (GEN-2755)
- Stripped vendor branding from tool descriptions
- Added
gen_run_research, expanded API_REFERENCE with agentic chat docs - Content monitoring + publishing tools
License
MIT
Comments
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.