MCP/Tellers.AI

Tellers.AI

Prompt-based video editing and AI video creation MCP server for generating, editing, rendering, and sharing videos from plain text.

video-editing•AI Video Generation & Prompt-Based Editing

GitHub7

License:Commercial / Free Tier

OFFICIAL

@Tellers.AI

About this Server

Official Model Context Protocol server that brings Tellers AI video creation and prompt-based video editing into any MCP-compatible client including Claude Desktop, Cursor, and Windsurf. Describe the video you want in plain language — the server handles everything from video generation and footage editing to multi-track sequencing, text overlays, audio scoring, rendering to MP4, and generating public sharing links.

Real-time Interaction

Direct low-latency RPC protocol communication with structured JSON schemas.

Scoped Sandboxing

Restricted capability bounds to prevent accidental modifications or leakage.

Security & Best Practices

API Key Protection: Store your TELLERS_API_KEY in environment variables or client configuration files without committing to version control.

Explicit Public Sharing: Assets remain strictly private to your account until you explicitly call tellers_makeAssetPublic.

Safe Local Media Access: Media uploads via tellers_uploadMedia access only explicitly specified local paths within user-authorized directories.

Session Isolation: Chat IDs isolate timeline modifications to their specific project sessions, preventing accidental timeline overwrites.

When to Use (and When Not To)

Best For

•Generating complete videos from scratch using natural language descriptions with AI agents
•Editing existing video clips, adding text overlays, color grading, and timing cuts via conversational prompts
•Automating video production workflows, rendering high-definition MP4 files, and generating public preview links
•Iteratively refining video timelines across multiple chat turns by continuing previous session chat IDs

Not For / Limitations

•Real-time interactive frame-by-frame scrubbing without rendering latency (use desktop NLEs like Premiere or Resolve instead)
•Offline rendering without an internet connection or Tellers API credentials
•Generating 3D polygonal geometry or procedural shader node graphs (use Blender MCP instead)

Install & Setup

One-line terminal command

terminal command

$claude mcp add tellers -- node /path/to/mcp-server/build/index.js -e TELLERS_API_KEY=your_api_key_here

Exposed Tools & Endpoints

8 tools

Functions your AI agent can invoke via this Model Context Protocol server

tellers_generateVideo• callable tool

Create or edit a video from a natural-language prompt with model configuration and style parameters.

tellers_uploadMedia• callable tool

Upload a local audio, image, or video file to use as source footage in the project.

tellers_exportProject• callable tool

Render a video project timeline to a downloadable high-definition MP4 file.

tellers_makeAssetPublic• callable tool

Publish a rendered video asset or timeline and generate a shareable preview link.

tellers_listRecentAssets• callable tool

Browse recent video assets, uploaded media files, and completed exports.

tellers_listChats• callable tool

List recent video generation sessions and chat threads for continuity.

tellers_getChatMessages• callable tool

View the detailed prompt message history and step-by-step revisions of a video project session.

tellers_getSettings• callable tool

Inspect available AI video models, rendering profiles, resolution options, and account configuration.