Setup & Installation
Or with the ClawHub CLI, for registry-managed skill folders outside a full OpenClaw workspace:
Version History
Bump runtime pin to v0.2.5: gemini 3.1 pro model, TikTok share URL fix, deterministic tool policy, typing indicator fixes, personality preservation, Instagram fail-fast retries
What This Skill Does
Finds, compares, and analyzes social videos across YouTube, TikTok, Instagram, and Twitter/X by running a managed agent that returns real video links with relevance notes. Auto-bootstraps a pinned runtime from the video-sourcing-agent repo when not already installed. Supports slash-command invocation and free-form queries.
Returns verified video URLs instead of fabricated results or generic text summaries.
When to use it
- Finding trending videos on a topic across YouTube and TikTok
- Comparing how different creators cover the same subject
- Researching how a brand appears in social video content
- Locating video examples to support a presentation or report
- Tracking content trends for a specific niche or keyword
Example Workflow
Here's how your AI assistant might use this skill in practice.
User asks: /video_sourcing find recent YouTube videos explaining transformer architecture
- 1Strips the '/video_sourcing' prefix and extracts the query text
- 2Builds the shell command with '--query', '--event-detail compact', and '--ux-mode three_message'
- 3Executes run_video_query.sh in the background with a 420-second timeout
- 4Polls for NDJSON output every 2-4 seconds and sends progress updates as they arrive
- 5Parses the terminal 'complete' event and formats the top 3 video results
Short conclusion paragraph followed by up to 3 videos, each with title, URL, and a one-line relevance note
Requirements
Accounts, API keys, or tools you or your AI assistant may need to set up while using this skill.