MCP.so
登录

Treza MCP Server已验证已精选

@Treza Labs

基本信息

传输方式

Streamable HTTP

认证方式

OAuth

分类

媒体与设计

作者

Treza Labs

提交者

Alex Daro

关于 Treza MCP Server

Lets Claude, ChatGPT and other MCP clients build and run AI video pipelines on Treza, from script and generated shots to narration, captions and posting to YouTube or TikTok.

连接信息

https://www.trezalabs.com/api/mcp

接入方式

claude mcp add treza --transport http https://www.trezalabs.com/api/mcp

Treza MCP Server 工具

工具列表暂未提供

当前收录暂时没有可展示的工具列表,不代表服务没有工具。请查阅项目文档了解支持的能力。

此服务需要认证,获取工具列表可能需要登录或使用服务提供方的凭据。

点击上方按钮,直接连接服务器端点实时获取工具列表。

概览

What is Treza

Treza is an AI video pipeline platform. Connect it to Claude, ChatGPT, Cursor or any MCP client and the assistant can build a video pipeline from a short brief, run it, and hand back the finished video. A pipeline can write the script, generate shots with models like Veo 3.1, Seedance, Wan, Kling and Gemini image, add narration, music and captions, and post the result to YouTube or TikTok.

The assistant can also finish media you already have without building a pipeline, for example joining clips into one video over a voiceover and a music bed, or captioning, upscaling or trimming a single file.

What you can ask for

  • Turn a brief into a captioned Short
  • Clip a long podcast or video into vertical clips on a schedule
  • Run an existing pipeline and get the finished video back
  • Stitch clips from your library into one captioned video with music
  • Estimate what a run will cost before spending anything
  • Pause or resume a channel's posting schedule

Tools

  • 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
  • Finishing media: list_assets, assemble_video, edit_asset, import_media, create_upload_url
  • Budgeting: estimate_run_cost, get_credit_balance
  • Publishing and access: list_connected_channels, connect_channel, publish_asset, create_api_key

Read-only tools are annotated as read-only, and tools that spend credits or post to a channel are marked so clients can ask before calling them.

Connecting

Add https://www.trezalabs.com/api/mcp to your client. It signs you in with OAuth, and signing in with Google creates a Treza account if you don't have one yet. Headless clients can use a scoped Treza API key as a bearer token instead.

Pricing

Prepaid credits, no subscription. Credit packs start at $5, and your first top-up gets 500 free credits. Publishing needs a YouTube or TikTok channel connected in the Treza app, and everything the assistant creates stays editable in the Treza editor.

Links

常见问题

1

Treza 远程 MCP 服务器是什么?

Treza 的远程 MCP 服务器是一个托管的 Model Context Protocol 端点,地址是 https://www.trezalabs.com/api/mcp,AI 助手无需在本地安装或运行任何东西即可连接。

2

怎么连接 Treza 的 MCP 服务器?

把端点 https://www.trezalabs.com/api/mcp 添加到任意支持 MCP 的客户端,比如 Claude Code、Cursor 或 VS Code。本页的接入代码块可以一步配置好每个客户端。

3

Treza 的 MCP 服务器需要认证吗?

需要。Treza 使用 OAuth——第一次连接时,你的 MCP 客户端会打开浏览器窗口登录并授权,之后会复用这个凭证。

4

Treza 的 MCP 服务器用的是哪种传输方式?

Treza 暴露的是 Streamable HTTP 端点,这是远程 MCP 服务器普遍使用、主流客户端都支持的传输方式。

评论