Treza MCP Server
io.github.treza-labs/treza
Build, run, schedule, and publish AI video pipelines to YouTube and TikTok from any MCP client.
What is the Treza MCP server?
The Treza MCP server enables AI agents to build, run, schedule, and publish video pipelines directly to YouTube and TikTok. It provides 24 tools across authoring, execution, media management, channel publishing, and credit budgeting, with support for video assembly, editing, and background rendering.
Treza lets you automate video creation and publishing workflows through AI agents. You can design video pipelines with templates and node types, manage a media library, connect YouTube and TikTok channels, estimate costs before running jobs, and publish finished videos—all from Claude, Cursor, or any MCP-compatible client. Rendering and publishing happen in the background and cost prepaid credits.
How to install Treza
Copy-paste configuration for popular MCP clients.
Tools & capabilities
Tools this server exposes to the agent.
list_pipeline_templates— Browse available video pipeline templateslist_node_types— List available node types for pipeline constructionget_node_type— Get details about a specific node typecreate_pipeline— Create a new video pipelineupdate_pipeline— Modify an existing pipelinepublish_pipeline— Publish a pipeline (spends credits)list_pipelines— List all pipelinesget_pipeline— Get details about a specific pipelinerun_pipeline— Execute a pipeline (spends credits, can post to connected channels)get_run— Get status of a pipeline runlist_runs— List all pipeline runsset_schedule_paused— Pause or resume a scheduled pipelinelist_assets— Browse the media libraryimport_media— Add a file to the media library from attachment, public link, or local filecreate_upload_url— Generate URL for uploading local filesassemble_video— Join clips into one video with optional narration, music, and captions (spends credits, renders in background)edit_asset— Run node operations on a single file: captions, upscale, trim, crop, etc. (spends credits, renders in background)list_connected_channels— List connected YouTube and TikTok accountsconnect_channel— Return a link to connect a YouTube channel or TikTok accountpublish_asset— Prepare a finished video for review before posting
Use cases
- Automatically generate and publish short-form videos to TikTok and YouTube on a schedule
- Build multi-step video editing workflows (upscale, caption, trim, assemble) and run them at scale
- Create AI-narrated video compilations from a media library with background music and burned-in captions
- Estimate production costs before rendering expensive video pipelines
- Connect multiple social channels and publish finished videos for review before they go live
Treza MCP server FAQ
Treza is an MCP server that lets AI agents build, run, schedule, and publish video pipelines to YouTube and TikTok. It provides 24 tools for pipeline authoring, execution, media management, channel publishing, and credit budgeting.
Treza uses a prepaid credit system. Video rendering and publishing consume credits. You can estimate costs before running jobs and top up your wallet through x402 payment integration.
Add this to `.cursor/mcp.json`: `{"mcpServers": {"treza": {"url": "https://www.trezalabs.com/api/mcp"}}}`. For Claude or other clients, use the remote endpoint with OAuth.
No. The recommended approach is to point your client at the remote endpoint `https://www.trezalabs.com/api/mcp`. For clients that only support stdio, use the `treza-mcp` Node bridge with a scoped API key.
Remote connections use OAuth with Google sign-in (creates a Treza account on first connection). Stdio connections require a scoped API key set in the `TREZA_API_KEY` environment variable.
Yes. Pipelines ending in `youtube-upload` or `tiktok-upload` nodes publish directly. Alternatively, use `publish_asset` to prepare videos for review before posting.
README (reference)
Source of truth, from the repository.
Treza plugin
Agent Plugins package for the Treza MCP server. Install it in Cursor, Codex, GitHub Copilot, VS Code, or any client that supports the standard, and your agent can build, run, schedule, and publish AI video pipelines to YouTube and TikTok.
- Server:
https://www.trezalabs.com/api/mcp(Streamable HTTP, OAuth with dynamic client registration) - Landing page and setup snippets: https://www.trezalabs.com/connect
- Docs: https://docs.trezalabs.com/api/mcp-server
- Privacy: https://docs.trezalabs.com/terms-and-privacy/privacy
Two ways to connect
Remote (recommended). Point any client that supports OAuth at https://www.trezalabs.com/api/mcp. The first connection signs you in with Google and creates the Treza account. Nothing to install.
stdio bridge. For clients and agent runtimes that only spawn local servers, this repo ships treza-mcp, a small Node bridge that forwards every tool, prompt, and resource to the hosted endpoint. It needs a scoped API key in TREZA_API_KEY (mint one in the Treza app under Settings, API keys, or have an OAuth-connected client call create_api_key).
{
"mcpServers": {
"treza": {
"command": "npx",
"args": ["-y", "treza-mcp"],
"env": { "TREZA_API_KEY": "treza_live_..." }
}
}
}
TREZA_MCP_URL overrides the endpoint (staging, self-hosted). The bridge runs nothing locally; it is a proxy.
What is inside
| File | Purpose |
|---|---|
bin/treza-mcp.mjs | The stdio bridge (source of the treza-mcp command) |
plugin.json | Agent Plugins manifest |
mcp.json | Declares the remote Treza MCP server for plugin-aware clients |
skills/treza-video-pipelines/ | A skill that teaches the agent the credits-first, estimate-before-run workflow |
Tools
Twenty-four tools in five groups:
- Authoring:
list_pipeline_templates,list_node_types,get_node_type,create_pipeline,update_pipeline,publish_pipeline - Running:
list_pipelines,get_pipeline,run_pipeline,get_run,list_runs,set_schedule_paused - Your media:
list_assetsbrowses the media library,import_mediaadds a file (an attachment, a public link, or a local file sent throughcreate_upload_url),assemble_videojoins clips into one video with optional narration, a music bed, and burned-in captions, andedit_assetruns node operations (captions, upscale, trim, crop, and more) over a single file - Channels and publishing:
list_connected_channels,connect_channelreturns a link that connects a YouTube channel or TikTok account in the person's own browser, andpublish_assetlays a finished video out for review before it posts.confirm_publishbelongs to the publish panel: it runs when the person presses Post there, never from the model - Budget and access:
estimate_run_cost,get_credit_balance,create_api_key(OAuth connections only)
Read-only tools carry readOnlyHint. run_pipeline, publish_pipeline, assemble_video, and edit_asset spend prepaid credits, and run_pipeline and publish_pipeline can post to a connected channel. assemble_video and edit_asset render in the background and return a run to poll with get_run.
In clients that show MCP Apps panels, such as ChatGPT and Claude, results appear as panels: the finished file, the price with a Start render button, a library to pick files from, a map of a pipeline, and the post before it goes out. Other clients get the same answers as text. Posting with publish_asset needs a panel; without one, publish through a pipeline that ends in a youtube-upload or tiktok-upload node.
Cursor without the plugin
Add to .cursor/mcp.json:
{
"mcpServers": {
"treza": { "url": "https://www.trezalabs.com/api/mcp" }
}
}
Runs spend prepaid credits on your Treza account. Agents with a wallet can top up over x402 with no human in the loop.
MIT licensed. Support: hello@trezalabs.com
Related MCP servers

Tribal
Self-hosted semantic memory over MCP for engineering teams to preserve tacit knowledge and reasoning.
UK company register lookups plus US, EU and UN sanctions screening. Every answer cites source.
Official FDA recalls, drug labels, adverse events and clinical trials, with cited sources.
Official SEC EDGAR filings, financial data and full-text search. Every answer cites its source.
Execute KQL queries against Microsoft Defender Advanced Hunting via natural language.
Execute KQL queries against Microsoft Defender Advanced Hunting via natural language.