spimov-mcp
Spimov MCP server (spimov-mcp PyPI package)
Documentation
spimov-mcp
MCP server for Spimov — dub videos via Claude Desktop / Claude Code / any MCP client.
Two transports:
- stdio — runs locally, spawned by your MCP client, authenticated with `SPIMOV_API_KEY` env var. Supports local file upload (`create_dub`).
- HTTP — long-lived server (we host at `mcp.spimov.com`), authenticated per-request with `Authorization: Bearer spk_live_...` (or `?api_key=`). Exposes both Streamable HTTP (`/mcp`, current spec) and legacy SSE (`/sse`). URL-only — no local file upload.
Both transports are thin wrappers over the public REST API at `https://spimov.com/api/v1`. The same API key, the same daily request quota.
1. Get an API key
Sign in at https://spimov.com → Settings → API keys → Create new key. Copy the `spk_live_...` token shown once.
2a. Use via stdio (Claude Desktop)
Install:
pip install spimov-mcpAdd to `~/Library/Application Support/Claude/claude_desktop_config.json` (macOS) or `%APPDATA%\Claude\claude_desktop_config.json` (Windows):
{
"mcpServers": {
"spimov": {
"command": "spimov-mcp",
"env": {
"SPIMOV_API_KEY": "spk_live_xxxxxxxxxxxxxxxxxxxxxxxx"
}
}
}
}Restart Claude Desktop. The "spimov" MCP server should appear; ask Claude to "dub /Users/me/Movies/clip.mp4 into Turkish using Spimov".
2b. Use via HTTP (Claude Code / Claude web / other clients)
Self-hosted: `MCP_PORT=8001 spimov-mcp-http`. Hosted endpoints:
- Streamable HTTP (recommended): `https://mcp.spimov.com/mcp`
- Legacy SSE: `https://mcp.spimov.com/sse`
Claude Code / config-file clients (header auth):
{
"mcpServers": {
"spimov": {
"url": "https://mcp.spimov.com/mcp",
"headers": {
"Authorization": "Bearer spk_live_xxxxxxxxxxxxxxxxxxxxxxxx"
}
}
}
}Claude web (claude.ai → Settings → Connectors → *Add custom connector*): the
UI only takes a URL, so put the key in the query string:
https://mcp.spimov.com/mcp?api_key=spk_live_xxxxxxxxxxxxxxxxxxxxxxxx> The key travels in the URL (stored in your connector config). Treat it like a
> password and rotate it from Settings → API keys if it leaks. Local file
> tools (`create_dub`, `download_video`) are unavailable here — use `dub_youtube`
> to dub, and `get_download_url` to get a clickable browser download link for the
> finished MP4.
3. Tools
| Tool | Purpose |
|---|---|
| `create_dub` | Upload a local file and queue a dubbing job (stdio only). |
| `dub_youtube` | Dub a YouTube URL server-side; optional auto-upload to a channel. |
| `get_job_status` | Poll a job (queued / processing / review / completed / failed). |
| `list_jobs` | Recent jobs for the account. |
| `list_segments` | Transcript segments (text, speaker, emotion, timing) of a job. |
| `update_segment` | Edit one segment's text/emotion/speaker/skip and resynth. |
| `remix_video` | Re-render audio mix / subtitles / lipsync without re-dubbing. |
| `download_video` | Save the finished MP4 locally (stdio only). |
| `get_download_url` | Get a shareable, time-limited browser download link for the MP4 (use this over HTTP/web). |
| `get_subtitles` | Fetch SRT/VTT (any embedded language). |
| `upload_to_youtube` | Publish (or retry) a finished job to a connected channel. |
| `cancel_job` | Cancel/delete. |
| `list_languages` | Source + target language list (17 target languages). |
| `get_quota` | Current minute & request usage vs. plan limits. |
Dubbing tools (`create_dub`, `dub_youtube`) accept an optional(quality voice, default), `chatterbox` (emotion voice), or `elevenlabs` (paid).
4. Override the API base (dev)
SPIMOV_API_BASE=http://localhost:8000/api/v1 spimov-mcpFrequently asked questions
What is spimov-mcp?
spimov-mcp is Spimov MCP server (spimov-mcp PyPI package)
How do I install spimov-mcp?
Open the GitHub repository and follow its README. Most MCP servers are added to your client's MCP config, then called by your agent.
Is spimov-mcp open source?
Yes — it is hosted on GitHub at https://github.com/erdememinaga/spimov-mcp and has 1 stars.
Related MCP tools
Cognee is the open-source AI memory platform for agents. Give your AI agents persistent long-term memory across sessions with a self-hosted knowledge graph engine.
Automate browser based workflows with AI
Hindsight: Agent Memory That Learns
A privacy-first app that strips AI watermarks from content you own.
Agent framework and applications built upon Qwen>=3.0, featuring Function Calling, MCP, Code Interpreter, RAG, Chrome extension, etc.
The power of Claude Code / GeminiCLI / CodexCLI + [Gemini / OpenAI / OpenRouter / Azure / Grok / Ollama / Custom Model / All Of The Above] working as one.
Run your own MCP server? See who uses it and what to fix.
Measure it with TrackMCP