Update to KEEP Highlights

{
  "description": "Guide ChatGPT on discovering highlights, analyzing audiovisual dynamics, and trimming media clips using the KEEP API.",
  "included_files": [],
  "name": "get-started",
  "skill_md_contents": "---\nname: get-started\ndescription: Guide ChatGPT on discovering highlights, analyzing audiovisual dynamics, and trimming media clips using the KEEP API.\n---\n\n# KEEP Media Highlights AI Skill\n\nThis skill enables ChatGPT to interact with the **KEEP Media Processing Engine** (`https://api.skeepit.co`) to analyze media files, detect potential highlight timestamps, inspect acoustic and visual waveforms, and assemble selected intervals into clips.\n\n## Core Capabilities\n\n1. **Highlight Detection (`discover_highlights`)**:\n   - Analyzes video or audio media files.\n   - Computes audiovisual activity curves (optical motion and acoustic energy).\n   - Generates 96-bin waveform samples with activity scores.\n   - Returns suggested segment intervals `[[start, end], ...]`.\n\n2. **Media Clipping (`condense_media`)**:\n   - Trims and stitches specified time intervals into a single output media stream.\n   - Supports MP4, WebM, MP3, and WAV container formats.\n\n## Recommended Workflow\n\n1. **When a user shares a video or audio URL**:\n   - Invoke `discover_highlights(media_url)` to compute duration and suggested highlight intervals.\n   - Present a clear summary of detected segments with timestamps (e.g. `00:00 - 01:15`, `04:30 - 05:45`).\n\n2. **When a user asks to trim or condense the media**:\n   - Propose or confirm the intervals to preserve.\n   - Invoke `condense_media(media_url, segments)` with the selected `[[start, end], ...]` array.\n   - Provide the resulting clip information to the user.\n\n3. **Safety & Boundaries**:\n   - KEEP processes media streams provided by the user.\n   - Out-of-bounds timestamps (beyond media duration) will return an error response.\n"
}