Tools
MCP server for explainer videos
The Chasca MCP server lets Claude, Cursor or any Model Context Protocol client make and change explainer videos for you, in the middle of whatever you are already doing.
https://app.trychasca.com/api Connect from Claude or ChatGPT
Add Chasca as a connector and sign in: nothing to install and no API key. It works on every plan, and videos count the same as in the app (your free videos, then your plan’s minutes).
https://app.trychasca.com/mcp - Claude (claude.ai and Claude Desktop, on every Claude plan): open Customize → Connectors, then Add → Custom → Web. Paste the address, continue, and sign in to Chasca.
- Claude Code: run the command below, then
/mcpto sign in. - ChatGPT: Business and Enterprise workspaces can add it as a custom MCP server. Any other MCP client that supports remote servers and OAuth works the same way.
claude mcp add --transport http chasca https://app.trychasca.com/mcp To disconnect an assistant, open Settings → API and MCP in the app.
On your computer, with an API key
A small MCP server, published on npm as @chasca/mcp, that runs on your computer over stdio with Node 20
or later. It talks to the Chasca API with your key, so everything an agent does shows up in your account (marked as created via
MCP) and counts like an API call: Plus and above, minutes at 2x.
CHASCA_API_KEY: required. Create one in the app under Settings → API keys.CHASCA_API_URL: optional, defaults tohttps://app.trychasca.com/api.
Claude Code
Register the server once:
claude mcp add chasca \
-e CHASCA_API_URL=https://app.trychasca.com/api \
-e CHASCA_API_KEY=chk_live_... \
-- npx -y @chasca/mcp
Add --scope user to make it available in every project, and check it with claude mcp list or
/mcp inside Claude Code.
Claude Desktop
Add Chasca to claude_desktop_config.json (on macOS in
~/Library/Application Support/Claude/, on Windows in %APPDATA%\Claude\), then restart
Claude Desktop:
{
"mcpServers": {
"chasca": {
"command": "npx",
"args": ["-y", "@chasca/mcp"],
"env": {
"CHASCA_API_URL": "https://app.trychasca.com/api",
"CHASCA_API_KEY": "chk_live_..."
}
}
}
}
If Claude Desktop can’t find npx, use its full path as command: where npx on
Windows or which npx on macOS prints it.
Cursor and other clients
Any client that speaks MCP over stdio works the same way: run npx -y @chasca/mcp with
CHASCA_API_KEY set. In Cursor, add the same mcpServers entry to
.cursor/mcp.json in your project (or ~/.cursor/mcp.json for all projects).
Tools
| Tool | Inputs | What it does |
|---|---|---|
chasca_generate_explainer | input, length, aspect, voice, language, preset, hand, pronunciations, wait | Creates a video from a topic, a question or a whole document. Returns the queued record, or waits until it is ready when wait is true. |
chasca_get_explainer | id | Status, stage and progress; the video URL once ready; the editable scene graph and every version. |
chasca_change_explainer | id, change, wait | Changes one thing and keeps the rest exactly as it was. Subtitles, board style, music, watermark, logo, credit line and how a word is said make a free new version; a new voice or format makes a separate video from the same script and drawings, charged like a new one. |
chasca_list_explainers | limit | Your latest videos, newest first, with a one-line status summary. |
chasca_list_voices | language | Every narrator, with its tone and a preview. |
Generation runs in the background. Agents either poll chasca_get_explainer or pass
wait: true to block until the video is ready. Errors come back with a plain hint for the agent, such as
how to fix a missing key or when to retry.
Things to ask for
- “Make a 1-minute vertical explainer about how our refund policy works, from the attached policy text.”
- “Turn this README into a 2:00 explainer in the Blueprint style, narrated by Atlas.”
- “Make a vertical copy of my last Chasca video for Reels.”
- “Which Spanish voices can narrate for Mexico? Use one to remake yesterday’s video in Spanish.”