Model Context Protocol
Kanto Labs MCP server
Give Claude, Cursor, VS Code Copilot and any other MCP client 6 video and image tools: video frame extraction, video to LLM context, video scene detection, video to slides, video OCR and AI image upscaling.
Listed in the official MCP Registry as io.github.Akhiyugo/kanto-labs-mcp. npx installs and builds it straight from GitHub on first run; the one-click .mcpb bundle is on the GitHub releases page.
How billing works
No Kanto Labs account. Your Apify token, your cap.
Each tool call runs the matching actor on Apify with your own Apify API token, so the actor's pay-per-result price is billed to your Apify account (Apify's free plan includes monthly credit). There is no Kanto Labs account, key or subscription.
Every call carries a spending cap, maxTotalChargeUsd (default $1.00; change it with KANTO_MAX_CHARGE_USD or per call). Apify stops the run once it has charged that much, and each result states what the call cost. Failed or unreachable inputs are not charged.
Example prompts
Just ask
- “Summarize this webinar recording scene by scene: https://example.com/webinar.mp4”
- “Pull 8 evenly spaced thumbnails from https://example.com/demo.mp4 and pick the best one.”
- “How many shots does this ad have, and how fast is it cut? https://example.com/ad.mp4”
- “Upscale https://example.com/product-small.jpg 4x as a PNG.”
- “Read the burned-in subtitles of https://example.com/clip.mp4 and give me an SRT file.”
Setup
Install in your client
- Get an Apify API token: sign up at apify.com, then Settings → API & Integrations.
- Add the server to your client with one of the snippets below. Node.js 18.17+ is required;
npxdownloads the server on first run.
Claude Code
claude mcp add kanto-labs -e APIFY_TOKEN=apify_api_your_token_here -- npx -y github:kanto-labs/kanto-labs-mcpClaude Desktop
Settings → Developer → Edit Config, then add:
{
"mcpServers": {
"kanto-labs": {
"command": "npx",
"args": ["-y", "github:kanto-labs/kanto-labs-mcp"],
"env": { "APIFY_TOKEN": "apify_api_your_token_here" }
}
}
}Cursor
~/.cursor/mcp.json (global) or .cursor/mcp.json (project): the same JSON as Claude Desktop.
{
"mcpServers": {
"kanto-labs": {
"command": "npx",
"args": ["-y", "github:kanto-labs/kanto-labs-mcp"],
"env": { "APIFY_TOKEN": "apify_api_your_token_here" }
}
}
}VS Code (GitHub Copilot agent mode)
The token is asked for once and stored securely.
{
"inputs": [
{ "type": "promptString", "id": "apify_token", "description": "Apify API token", "password": true }
],
"servers": {
"kanto-labs": {
"type": "stdio",
"command": "npx",
"args": ["-y", "github:kanto-labs/kanto-labs-mcp"],
"env": { "APIFY_TOKEN": "${input:apify_token}" }
}
}
}Windsurf, Cline, Zed and others: any client that launches stdio servers works: command npx, args -y github:kanto-labs/kanto-labs-mcp, env APIFY_TOKEN.
Windows: if your client reports spawn npx ENOENT, use "command": "cmd" with "args": ["/c", "npx", "-y", "github:kanto-labs/kanto-labs-mcp"].
Tools
What the server exposes
| Tool | What it does | Price |
|---|---|---|
extract_video_frames | Frames from video files | $0.002 per frame |
video_to_llm_context | Scenes, keyframes and transcript per scene for an AI model | $0.002 per keyframe + $0.008 per audio minute |
detect_video_scenes | Shot list with cut timecodes, keyframes and an EDL | $0.018 per video minute ($0.002 per clip) |
upscale_image | AI 2x / 4x image upscaling | $0.01 per image ($0.06 large) |
video_to_slides | Unique slides of a lecture or webinar as images, text and a PDF | $0.004 per slide + $0.001 per video minute |
video_ocr | On-screen text and burned-in subtitles as timed segments and SRT | $0.002 per frame with text |
Privacy
The server runs on your machine and talks only to api.apify.com. Your token is sent only to Apify and is never logged.