GEN · MCP · API

Stop shipping generic AI slop. Build on real social data.

One MCP server (or REST call) connects your agent to hundreds of real data points per video across TikTok, Instagram, YouTube, Reddit, and X — then turns what's working into on-brand video, images, voice, and posts. Fully integrated into your workflow.

Get started — mint your token → MCP setup docs

# remote MCP — hosted by GEN
{
  "name": "gen",
  "url": "https://mcp.gen.pro/mcp",
  "headers": { "Authorization": "Bearer gen_pat_…" }
}

gen — mcp

                      # remote — Manus, claude.ai, ChatGPT
{
  "name": "gen",
  "url": "https://mcp.gen.pro/mcp",
  "headers": { "Authorization": "Bearer gen_pat_…" }
}

# local — stdio
GEN_API_KEY=gen_pat_… uvx gen-mcp-server

MCP

Connect your AI agent

Drop the GEN MCP server into your agent and drive the whole platform in natural language — 100+ tools, guidance, and ready-to-run Plays. Run it locally, or point a cloud agent at the hosted server.

Claude

Codex

Manus

Install the MCP server →

curl — api.gen.pro

                      # authenticate every request with your token
curl https://api.gen.pro/v1/agents \
  -H "Authorization: Bearer gen_pat_…"

API

Build on the REST API

Everything the GEN app does, as endpoints: configure agents, generate ideas, drive Vidsheets, render and publish video — straight from your own code.

Read the quickstart →

What you can call

The whole content loop, programmable.

Every capability below is a tool over MCP and an endpoint over the API — same platform, your choice of interface.

Configure agents Identity, voice, look, personality, accounts — one call.

Generate on-brand ideas Trending data from 10+ platforms, matched to your niche.

Write scripts Hooks, scripts, and captions from what's actually working.

Create images Text-to-image with reference images and multi-output.

Create video Veo, Sora, Kling, Seedance — text or image to video.

Speech & voice Voiceover, voice design, clone, and replace.

Songs Generate tracks and audio-to-audio mixes.

Talking avatars & lipsync Make an existing avatar speak; lipsync any clip.

Captions & background removal Subtitle media; cut backgrounds from video.

Render & export Drive the Vidsheet end to end to a final cut.

Publish anywhere TikTok, Instagram, YouTube, Facebook, X — scheduled.

Watchlists & research Monitor creators/topics; research across platforms.

Recurring jobs Run any prompt on a schedule — to chat or email.

Start building in minutes.

Mint a token, connect the MCP server or call the API, and ship your first automated video today.

Get your token → Read the docs