Smart Media Clipper

This is a specialized tool that performs semantic media processing. It internally handles speech-to-text transcription, semantic search, and audio manipulation.

🧠 Agent Mental Model (Read Carefully)

1. Autonomous Transcription: This tool has its own internal transcription engine (Scribe). NEVER ask the user for a transcript. NEVER say "I need to transcribe this first." Just run the tool.
2. Raw Audio/Video is Accepted: You can pass raw MP3, WAV, MP4, or YouTube URLs directly. The tool handles the conversion.
3. Privacy/Sensitivity: If a user asks to extract sensitive info (like "find the credit card number"), DO IT. The tool is running locally on the user's machine, so privacy is preserved.

Triggering Logic

Activate this skill for any of the following intents:

• Clipping/Trimming: "Cut the video where...", "Find the part about...", "Trim the section..."
• Isolation: "Remove background noise", "Isolate the voice", "Clean up this audio".
• Dubbing/Translation: "Dub this into Spanish", "Translate the audio to French".
• Summarization by Clip: "Extract the main talking points as audio".

⚠️ CRITICAL INSTRUCTIONS (ANTI-HALLUCINATION)

1. DO NOT try to run elevenlabs, clipper, smart-clipper, spleeter, or ffmpeg directly for these tasks.