Browse the docs

Preview video

Shows a video project before rendering it: up to 4 still frames (the opening, early, middle and late) and previewUrl, a page that plays it in the browser for 7 days without signing in. Free (no credits or minutes), limited to 20 a minute. Check the look here, change it with update_video, then render_video.

Or open inClaude Code Codex Cursor Claude ChatGPT

Paste it into any AI assistant. The shortcuts open with the prompt typed in; nothing is sent until you press enter.

At a glance

Property	Value
REST	`POST https://api.clippie.ai/v1/videos/:projectId/preview`
MCP tool	`preview_video`
Scope	`read`
Cost	Free. Works with a read-only key.
Limit bucket	`preview` (see Rate limits)

REST
POST https://api.clippie.ai/v1/videos/:projectId/preview
MCP tool
preview_video
Scope
read
Cost
Free. Works with a read-only key.
Limit bucket
preview (see Rate limits)

Parameters

Name	In	Type	Required	Default	Description
`projectId`	path	string	Yes		The project id: output.projectId of a job, an id from list_videos, or the id in an editor URL. 1-64 chars.
`template`	body	string	No		Only look among this template's projects (ai_story, captions, reddit, roblox). One of `ai_story`, `captions`, `reddit`, `roblox`.

projectId
In:
path
Type:
string
Required:
Yes
Description:
The project id: output.projectId of a job, an id from list_videos, or the id in an editor URL. 1-64 chars.
template
In:
body
Type:
string
Required:
No
Description:
Only look among this template's projects (ai_story, captions, reddit, roblox). One of ai_story, captions, reddit, roblox.

Example

curl

curl https://api.clippie.ai/v1/videos/PROJECT_ID/preview \
  -H "Authorization: Bearer $CLIPPIE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{}'

preview-video.mjs

const auth = { Authorization: `Bearer ${process.env.CLIPPIE_API_KEY}` };

const res = await fetch('https://api.clippie.ai/v1/videos/PROJECT_ID/preview', {
  method: 'POST',
  headers: { ...auth, 'Content-Type': 'application/json' },
  body: JSON.stringify({}),
});
if (!res.ok) throw new Error((await res.json()).error.message);
console.log(await res.json());

Response

200 with JSON. The full response schema is in the OpenAPI spec.

From an MCP client

Your AI calls preview_video with the same fields as JSON arguments. See MCP server to connect one.

tools/call

{
  "name": "preview_video",
  "arguments": {
    "projectId": "PROJECT_ID"
  }
}