Tegas MCP Server
OfficialClick 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., "@Tegas MCP ServerCreate a 15s video about a cozy coffee shop in Russian, show scenario and price."
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.
Tegas MCP Server
Tegas turns an idea into a finished short video. An agent writes a scenario with Tessa (our scenario model), optionally attaches reference images of your product or characters, generates a 5–60 second video with voice-over, subtitles and music, and can then edit the result — scenes, voice, subtitles, music, volumes — and rebuild it.
Tegas runs a hosted remote MCP server, so there is nothing to install or self-host:
https://api.tegas.ai/mcpTransport: Streamable HTTP (stateless)
Auth:
Authorization: Bearer tg_live_...Russian mirror:
https://api.tegasai.ru
This repository holds the public metadata and documentation for the Tegas MCP server (
server.json, install instructions, issue tracker). The server implementation itself is closed source.
Prerequisites
A Tegas account — sign up at tegas.ai (RU: tegasai.ru).
An active paid plan. Video rendering and hero sheets spend tokens, and the paid tools refuse to run without an active plan.
An API key: in the Tegas cabinet go to Settings → API → Create key. The key looks like
tg_live_.... Treat it like a password.
Related MCP server: ittybitty MCP server
Install
Claude Code
claude mcp add --transport http tegas https://api.tegas.ai/mcp \
--header "Authorization: Bearer tg_live_..."Cursor
Add to ~/.cursor/mcp.json (global) or .cursor/mcp.json (per project):
{
"mcpServers": {
"tegas": {
"url": "https://api.tegas.ai/mcp",
"headers": {
"Authorization": "Bearer tg_live_..."
}
}
}
}VS Code
Add to .vscode/mcp.json in your workspace (this keeps the key out of the file by prompting
for it once and storing it in VS Code's secret storage):
{
"inputs": [
{
"id": "tegas-key",
"type": "promptString",
"description": "Tegas API key (tg_live_...)",
"password": true
}
],
"servers": {
"tegas": {
"type": "http",
"url": "https://api.tegas.ai/mcp",
"headers": {
"Authorization": "Bearer ${input:tegas-key}"
}
}
}
}Claude.ai custom connector
Tegas is not (yet) in the Claude connectors directory, so add it as a custom connector:
Open Settings → Connectors on claude.ai.
Choose Add custom connector.
URL:
https://api.tegas.ai/mcp.Under advanced settings, add the header
Authorizationwith the valueBearer tg_live_....
Custom connectors are available on paid Claude plans. See Anthropic's guide to custom connectors.
Tools
12 tools. Only two of them ever spend tokens.
Tool | Cost | What it does |
| free | balance, plan, limits, price per second of video by quality |
| free | available voices for a language |
| free | uploads one image, returns a |
| free | uploads one track, returns a |
| free | Tessa writes a scenario from an idea |
| free | brief of characters / hero product, with the price per sheet |
| tokens | draws hero sheets — requires |
| tokens | starts the render — requires |
| free | status of one video |
| free | polls every 10 s until completed / failed / timeout |
| free | recent videos |
| free | edits scenes / voice / subtitles / music / volumes and rebuilds the video |
Paid tools never spend tokens on the first call. With confirm=false (the default) they
return the price and wait for your consent; only a second call with confirm=true charges.
Every render carries an Idempotency-Key, so retrying after a timeout never charges twice.
A failed render is refunded.
Current prices are always live from tegas_account — check there rather than trusting any
number written down here.
Example
Make a 15-second vertical video about a cozy coffee shop in Russian, 720p. Show me the scenario and the price first.
Privacy and data
The prompts, ideas, reference images and music you pass to the tools are sent to Tegas servers to generate and store your videos, and are associated with the account that owns the API key.
Reference images on the hosted connector must be passed as base64 — the connector cannot read files from your machine (
file_pathworks only on local/self-hosted runs withMCP_ALLOW_LOCAL_FILES=true).Your API key identifies and bills your account. Do not paste it into shared configs or commit it to a repository.
Full details: Tegas Privacy Policy (RU: tegasai.ru/privacy-policy) and the terms of offer.
Docs and support
Documentation: docs.tegas.ai
Agent-oriented quick reference: docs.tegas.ai/quickstart.md (also served to agents as the MCP resource
tegas://quickstart)OpenAPI schema: api.tegas.ai/v1/openapi.json
Bugs in this metadata or in the MCP server: open an issue
License
The contents of this repository — documentation and metadata only — are released under the MIT License. The MIT license covers these files alone. It does not apply to the Tegas MCP server, the Tegas API, or the Tegas service, which are proprietary and governed by the terms of offer.
This server cannot be deployed
Maintenance
Related MCP Connectors
Create and edit AI videos from chat: plan shots, generate scenes, and export stories and ads.
Generate finished AI videos from a brief or a script, with voiceover, scenes and captions.
Make videos and docs with your AI agent — describe what you need, every output stays editable.
Autonomous short-form video for brands your agent provisions — to your webhook, or posted for them.
Related MCP Servers
- AlicenseNot gradedqualityDmaintenanceEnables AI agents to edit videos through natural language, providing tools for timeline editing, audio management, rendering, and more.2MIT
- AlicenseNot gradedqualityAmaintenanceEnables AI agents to generate narrated videos from topics or scripts, with stock footage, home videos, or local AI clips.2MIT
- AlicenseAqualityCmaintenanceEnables AI agents to edit video assemblies from A-roll and B-roll, add captions, and publish to social media platforms.279 npmMIT
- AlicenseNot gradedqualityBmaintenanceEnables AI agents to produce professional promo videos by providing tools for capture, media search, scene writing, rendering, editing, and incremental revision.14 npm15MIT