seedance-polza-mcp-server

# Seedance Polza MCP Server `mcp-name: io.github.ivanantigravity-lgtm/seedance-polza-mcp-server` MCP server for generating video through `bytedance/seedance-2` (and other Seedance models) on Polza.ai. ## What it can do - text-to-video and image-to-video generation - reading generation status - polling until video is ready - returns compact result (id, status, url, usage, warnings) ## Requirements for installation - `Claude Desktop` or `Claude Code` - [`uv`](https://docs.astral.sh/uv/) - Python 3.11+ - `POLZA_AI_API_KEY` — key taken from [polza.ai/dashboard/api-keys](https://polza.ai/dashboard/api-keys) Install `uv`: ```bash curl -LsSf https://astral.sh/uv/install.sh | sh ``` ## Installation in 2 minutes (via PyPI + uvx) ### Claude Code / VS Code Create `.mcp.json` in project root: ```json { "mcpServers": { "seedance-polza": { "command": "uvx", "args": ["seedance-polza-mcp-server@latest"], "env": { "POLZA_AI_API_KEY": "your-polza-api-key-here" } } } } ``` Restart Claude Code. ### Claude Desktop (macOS) File `~/Library/Application Support/Claude/claude_desktop_config.json`: ```json { "mcpServers": { "seedance-polza": { "command": "uvx", "args": ["seedance-polza-mcp-server@latest"], "env": { "POLZA_AI_API_KEY": "your-polza-api-key-here" } } } } ``` ### Claude Desktop (Windows) File: `%APPDATA%\Claude\claude_desktop_config.json`. Contents are identical. ## How to check if it works After restarting Claude, ask: > Make a short video through seedance: sunset over the ocean, 16:9, 5 seconds Claude should call `seedance_create_video` and wait for completion via `seedance_wait_for_completion`. ## Tools - `seedance_create_video` — start generation - `seedance_get_status` — check status by `id` - `seedance_wait_for_completion` — wait by polling until ready - `seedance_model_guide` — brief guide on parameters ## Defaults - model: `bytedance/seedance-2` - polling interval: `8` seconds - max wait: `900` seconds ## Environment Variables | Variable | Required | Default | | --- | --- | --- | | `POLZA_AI_API_KEY` | yes | — | | `POLZA_BASE_URL` | no | `https://polza.ai/api/v1` | | `SEEDANCE_MODEL` | no | `bytedance/seedance-2` | | `SEEDANCE_POLL_INTERVAL` | no | `8` | | `SEEDANCE_MAX_WAIT` | no | `900` | | `LOG_LEVEL` | no | `INFO` | ## Supported input parameters Basic parameters for video: - `prompt` - `aspect_ratio` (`16:9`, `9:16`, `1:1`, `4:3`, `3:4`, `21:9` etc.) - `resolution` (`480p`, `720p`, `1080p`) - `duration` (`5s`, `10s`, `15s` — depends on model) - `images` — references for image-to-video - `videos` — references for video-to-video - `seed` - `async` - `user` Format `images` / `videos`: ```json { "type": "url", "data": "https://example.com/file.png" } ``` or ```json { "type": "base64", "data": "data:image/png;base64,..." } ``` ## What the server returns The server does not perform extra magic. It returns: - `id` generation - `status` - `model` - `usage` (including `cost_rub`) - `url` result, if video is ready - `warnings` / `error` in case of problems ## Local development ```bash git clone https://github.com/ivanantigravity-lgtm/seedance-polza-mcp-server.git cd seedance-polza-mcp-server uv sync POLZA_AI_API_KEY=your_key uv run python -m seedance_polza_mcp_server.server ``` ## Important This server does not run live preview and does not spend tokens itself. Generation is started only by explicit tool call. ## License MIT.