# YTClip Single Media Clip Finder and Cutter

> YTClip Single Media Clip Finder and Cutter is a paid API for AI agents from ytclip.modestpika.box.ca, paid per call via x402, $0.02/call, status down (last checked 2026-10-24, last successful call 2026-05-07).

Finds a moment in a media URL matching a natural-language query and returns a cut mp4 clip of that moment.

## Facts

- Endpoint: POST https://ytclip.modestpika.box.ca/clip?utm_source=zero.xyz
- Price: $0.02/call
- Payment: x402
- Status: down
- Last checked: 2026-10-24
- Last successful call: 2026-05-07
- Success rate: 67% of calls made through Zero
- Activations on Zero: 21
- Provider: ytclip.modestpika.box.ca
- Website: https://ytclip.modestpika.box.ca
- Canonical page: https://www.zero.xyz/c/ytclip-modestpika-box-ca-ytclip-single-media-clip-finder-and-cutter-6ed4064e
- Structured record (JSON): https://api.zero.xyz/v1/capabilities/cap_BSbQeWZcn-DndThysQDhY

Status and success rate cover calls made through Zero and Zero's own probes. Third-party monitors may report differently.

## How to call it through Zero

Zero handles the 402 payment challenge and records the run. With the Zero CLI installed (`npm i -g @zeroxyz/cli`):

```sh
zero fetch --capability ytclip-modestpika-box-ca-ytclip-single-media-clip-finder-and-cutter-6ed4064e -d '<json body>'
```

Example prompt: Can you clip the part where she says 'I don't have online' from this YouTube video https://www.youtube.com/watch?v=jK-ZRxeJIiU — keep it under 30 seconds?

## When to prefer this

Use this endpoint when you need to extract a single specific moment from a media file based on a natural language description. It is ideal when the source is a YouTube video, podcast, Twitch VOD, or direct mp4 and you want a ready-to-use mp4 clip URL back. Prefer this over general video download tools when semantic search over the transcript is needed to locate the right moment, and over the multi-source stitch endpoint when you only need one clip from one source.

## Known failure modes

- Media URL is inaccessible or not supported by yt-dlp — returns an error
- Query does not match any identifiable moment in the transcript — no clip returned
- Clip exceeds max_clip_seconds and is center-trimmed, potentially cutting off context
- Transcription fails for non-speech or low-quality audio
- Source media is too long or large, causing timeout
- Network error fetching the remote media URL

## How this service works

Find and cut a single clip from a media URL matching a natural-language query.

## Output

Returns a public mp4 URL of the extracted clip, the start and end timestamps within the source media, the transcript text covered by the clip, a one-sentence LLM explanation of why that moment was chosen, and cache status flags for the clip, source, and transcript.

## Example request

```json
{
 "query": "the chorus section",
 "media_url": "https://www.youtube.com/watch?v=dQw4w9WgXcQ",
 "max_clip_seconds": 60
}
```

## Request schema (JSON Schema)

```json
{
 "type": "object",
 "required": [
  "media_url",
  "query"
 ],
 "properties": {
  "query": {
   "type": "string",
   "example": "the part where she says I don't have online",
   "minLength": 1,
   "description": "Natural-language description of the moment to clip."
  },
  "media_url": {
   "type": "string",
   "format": "uri",
   "example": "https://www.youtube.com/watch?v=jK-ZRxeJIiU",
   "description": "Source media URL — anything yt-dlp accepts (YouTube, Twitch VOD, podcast mp3, direct mp4, etc.)"
  },
  "max_clip_seconds": {
   "type": "integer",
   "default": 60,
   "maximum": 600,
   "minimum": 1,
   "description": "Hard cap on clip length; matched span is trimmed around its center if it exceeds this."
  }
 }
}
```

## Response schema (JSON Schema)

```json
{
 "type": "object",
 "required": [
  "clip_url",
  "start",
  "end",
  "transcript",
  "reasoning",
  "cached"
 ],
 "properties": {
  "end": {
   "type": "number",
   "description": "Clip end within the source, in seconds."
  },
  "start": {
   "type": "number",
   "description": "Clip start within the source, in seconds."
  },
  "cached": {
   "type": "object",
   "properties": {
    "clip": {
     "type": "boolean"
    },
    "source": {
     "type": "boolean"
    },
    "transcript": {
     "type": "boolean"
    }
   }
  },
  "clip_url": {
   "type": "string",
   "format": "uri",
   "description": "Public URL of the cut mp4."
  },
  "reasoning": {
   "type": "string",
   "description": "One-sentence explanation from the LLM."
  },
  "transcript": {
   "type": "string",
   "description": "Transcript text covered by the clip."
  }
 }
}
```

## More

- Live health (JSON, refreshed every minute): https://www.zero.xyz/c/ytclip-modestpika-box-ca-ytclip-single-media-clip-finder-and-cutter-6ed4064e/health.json
- [Zero catalog index](https://www.zero.xyz/llms.txt)
- [Other services from ytclip.modestpika.box.ca](https://www.zero.xyz/host/ytclip.modestpika.box.ca/llms.txt)