elevenlabslisted

# ElevenLabs Use this plugin when the user explicitly asks for ElevenLabs or when audio/speech quality matters enough to choose ElevenLabs. Do not title the workflow as generic transcription. The product surface is ElevenLabs; transcription is one documented use case through Scribe. ## Scribe Use Case Use the bundled CLI: ```bash python3 scripts/transcribe_elevenlabs.py \ meeting.mp4 \ --language es \ --diarize \ --num-speakers 2 \ --response-format diarized_text \ --out output/elevenlabs/meeting/transcript.txt ``` Resolve `scripts/transcribe_elevenlabs.py` relative to the ElevenLabs plugin root. If you are reading this skill from a plugin cache path, the helper is two directories above this skill file at `../../scripts/transcribe_elevenlabs.py`. ## Decision Rules - Default to `scribe_v2` for Scribe. - For meetings, prefer `--diarize --num-speakers 2` when the user knows it is a two-speaker conversation. - Use `--response-format diarized_text` when the goal is a readable speaker transcript. - Use `--response-format json` when raw API evidence matters. - Add `--language es` when the recording is known to be Spanish. - Add `--keyterm` for domain jargon the model is likely to miss. - Leave verbatim mode on unless the user wants filler words and false starts removed. ## Environment - `ELEVENLABS_API_KEY` must be set for live API calls. - Never ask the user to paste the full key in chat. ## Reference Map - `references/api.md`: Scribe request knobs, limits,