Clip.A.Canvas MCP server — render HTML/CSS/JS animations to MP4 via MCP tools
Project description
Clip.A.Canvas MCP Server
A universal MCP (Model Context Protocol) server that exposes Clip.A.Canvas HTML-to-video rendering as an MCP tool. Works with any STDIO-based MCP client.
Features
- Universal MCP tool —
render_videoandrender_video_to_fileexposed via the MCP protocol - Works everywhere — Claude Desktop, Codex, Gemini CLI, Qwen Coder, and any MCP-compatible AI coding tool
- Local rendering — no upload, no cloud; Chromium + FFmpeg run on your machine
- Base64 or file output — return video as base64 string or save directly to a path
Installation
One-command MCP install
After publishing to PyPI:
# Gemini CLI
gemini mcp add clipacanvas -- uvx clipacanvas-mcp
# Codex CLI
codex mcp add clipacanvas -- uvx clipacanvas-mcp
# Claude Code on Windows
claude mcp add -s user clipacanvas -- cmd /c uvx clipacanvas-mcp
From PyPI
pip install clipacanvas-mcp
# or
uv tool install clipacanvas-mcp # gives 'clipacanvas-mcp' command
From source (editable)
cd code2video/mcp
pip install -e .
uvx handles the Python package environment. On first render, the MCP package will also install Playwright Chromium automatically if it is missing.
Configuration
Environment variables (all optional):
| Variable | Default | Description |
|---|---|---|
CLIPACANVAS_FFMPEG_EXE |
auto-detect | Path to FFmpeg executable |
CLIPACANVAS_BROWSERS_PATH |
auto-detect | Path to Playwright browsers |
CLIPACANVAS_MAX_DURATION |
12 |
Max render duration in seconds |
CLIPACANVAS_DEFAULT_WIDTH |
540 |
Default video width |
CLIPACANVAS_DEFAULT_HEIGHT |
960 |
Default video height |
CLIPACANVAS_DEFAULT_BITRATE |
5M |
Default FFmpeg bitrate |
CLIPACANVAS_DEFAULT_FRAME_RATE |
60 |
Default frame rate |
MCP Tools
render_video
Render HTML/CSS/JS to MP4, returned as base64.
Arguments:
{
"code": "<!DOCTYPE html><html>...</html>",
"width": 540,
"height": 960,
"bitrate": "5M",
"frame_rate": 60,
"max_duration": 12
}
Returns: Text content with video metadata and base64-encoded MP4 data.
render_video_to_file
Same as above but saves the MP4 directly to output_path.
Arguments:
{
"code": "<!DOCTYPE html><html>...</html>",
"output_path": "/absolute/path/to/video.mp4",
"width": 540,
"height": 960
}
Usage in AI Coding Tools
Claude Code
For this local repo checkout:
cd code2video/mcp
pip install -e .
claude mcp add -s local clipacanvas -- cmd /c python -m clipacanvas_mcp.server
The cmd /c wrapper is useful on Windows because Claude Code's MCP health check can fail to resolve the same Python command that works in an interactive PowerShell session.
Claude Desktop
Add to your Claude Desktop config file:
macOS: ~/Library/Application Support/Claude/claude_desktop_config.json
Windows: %APPDATA%\Claude\claude_desktop_config.json
{
"mcpServers": {
"clipacanvas": {
"command": "python",
"args": ["-m", "clipacanvas_mcp.server"]
}
}
}
Then restart Claude Desktop. The render_video and render_video_to_file tools will appear in the tool list.
Codex CLI
For this local repo checkout:
cd code2video/mcp
pip install -e .
codex mcp add clipacanvas -- python -m clipacanvas_mcp.server
Gemini CLI (Google)
For this local repo checkout, install the server in editable mode first:
cd code2video/mcp
pip install -e .
gemini mcp add clipacanvas -- python -m clipacanvas_mcp.server
Use the uvx command only after clipacanvas-mcp is published to a package index and uvx is installed:
gemini mcp add clipacanvas -- uvx --from clipacanvas-mcp clipacanvas-mcp
Qwen Coder (Alibaba)
Add to Qwen Coder's MCP configuration:
{
"mcpServers": {
"clipacanvas": {
"command": "uvx",
"args": ["--from", "clipacanvas-mcp", "clipacanvas-mcp"]
}
}
}
Any STDIO MCP Client
# Direct run
uvx --from clipacanvas-mcp clipacanvas-mcp
# Or after pip install
clipacanvas-mcp
Testing the Server Manually
# Start the server
python -m clipacanvas_mcp.server
# Send initialize
echo '{"jsonrpc":"2.0","id":1,"method":"initialize","params":{"protocolVersion":"2024-11-05","capabilities":{},"clientInfo":{"name":"test","version":"1.0"}}}' | clipacanvas-mcp
# List tools
echo '{"jsonrpc":"2.0","id":2,"method":"tools/list","params":{}}' | clipacanvas-mcp
Requirements
- Python 3.10+
- FFmpeg (installed system-wide, via
imageio-ffmpeg, or setCLIPACANVAS_FFMPEG_EXE) - Playwright Chromium: installed automatically on first render if missing; run
python -m playwright install chromiummanually only if you want to preinstall it
License
MIT
Release history Release notifications | RSS feed
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.
Source Distribution
Built Distribution
Filter files by name, interpreter, ABI, and platform.
If you're not sure about the file name format, learn more about wheel file names.
Copy a direct link to the current filters
File details
Details for the file clipacanvas_mcp-1.0.2.tar.gz.
File metadata
- Download URL: clipacanvas_mcp-1.0.2.tar.gz
- Upload date:
- Size: 9.4 kB
- Tags: Source
- Uploaded using Trusted Publishing? No
- Uploaded via: twine/6.2.0 CPython/3.12.10
File hashes
| Algorithm | Hash digest | |
|---|---|---|
| SHA256 |
4bdd9bc7b347bd5723abef5dfff930622972946df30c20f28b76924d97bade25
|
|
| MD5 |
df7693c3b6311af1e53e330ffb5a5740
|
|
| BLAKE2b-256 |
59e5eaf6f00b70c2b4f266eec191164e84e77779b1b85342524805301e1af926
|
File details
Details for the file clipacanvas_mcp-1.0.2-py3-none-any.whl.
File metadata
- Download URL: clipacanvas_mcp-1.0.2-py3-none-any.whl
- Upload date:
- Size: 11.7 kB
- Tags: Python 3
- Uploaded using Trusted Publishing? No
- Uploaded via: twine/6.2.0 CPython/3.12.10
File hashes
| Algorithm | Hash digest | |
|---|---|---|
| SHA256 |
70132a8bb2c5cca3e7218fc496527a9d6d90b28e232ef0697b5748f4c8119785
|
|
| MD5 |
97315bf583a83255a8b476a9c2ee3dcb
|
|
| BLAKE2b-256 |
de7068125664108f40ab2195fb6548b84bdf2b8c0913ef519180d1d598c0753f
|
