Highlight Clip Extractor

Identify high-engagement moments in long-form recordings and extract standalone short-form video clips.

When to use

• Extracting multiple standalone highlight clips from a long lecture, interview, or stream.
• Selecting segments based on transcript hooks and narrative peaks.
• Batch-exporting short clips with vertical reframing and captions.

Tool interface

Use the registered ai-video-editor (OpenClaw plugin) or autonomous_edit (MCP server) tool to run edits against a project timeline.

Tool parameters

• prompt (string, required): Description of the video edit to perform.
• project_id (string, optional): Project identifier returned by a prior call for multi-turn continuity.
• video_url (string, optional): Source video URL when starting a new project.
• captionTemplatePreset (string, optional): Built-in caption style slug (for example hormozi, viral-pop, karaoke, minimal-pro).
• captionTemplateMode (string, optional): One of add, recreate, style, translate, clear, remove_fillers.
• requirePlanApproval (boolean, optional): When true, pauses after planning and returns workingMemory for user confirmation.
• workingMemory (object, optional): State object from a paused plan-approval call.

Example tool input

{
  "prompt": "Extract 3 standalone 30-second highlight clips from the most engaging moments in this recording.",
  "project_id": "project_example_01"
}

Capabilities

• Transcript and narrative peak analysis for highlight selection.
• Batch subclip extraction with independent timelines.
• Optional 9:16 vertical reframing and captioning per extracted clip.
• Consolidated MP4 and archive delivery.

Response fields

• success (boolean): True when the edit and verification passed.
• status (string): One of completed, waiting, awaiting_approval, or failed.
• projectId (string): Active project identifier for subsequent edits.
• projectRevision (number): Committed project revision number.
• videoUrl (string, optional): Rendered MP4 download URL when export completes.
• jobId (string, optional): Async render job identifier for status polling via check_job_status.