Update to Traspion

{
  "description": "Transcribe an attached audio or video file with Traspion, including speech-to-text, audio-to-text, video-to-text, speaker detection, or translation requests. Use when the user wants a new recording converted into a transcript.",
  "included_files": [],
  "name": "traspion-transcribe-media",
  "skill_md_contents": "---\nname: traspion-transcribe-media\ndescription: Transcribe an attached audio or video file with Traspion, including speech-to-text, audio-to-text, video-to-text, speaker detection, or translation requests. Use when the user wants a new recording converted into a transcript.\n---\n\n# Transcribe audio or video to text with Traspion\n\nUse `upload_and_transcribe` once for the audio or video attachment the user identifies.\n\n- If no attachment is available, or multiple attachments make the target ambiguous, ask the user to identify one file.\n- If the user has not specified speaker handling, ask one combined question before uploading: “Would you like speaker labels? If yes, how many speakers are there (2–30), or should Traspion detect the number automatically?” Do not ask a second confirmation after the answer; the original transcription request already authorizes the upload.\n- If the answer is no, set `speakerPartitioning: false` and omit `speakersCount`. If the answer is yes with a count, set `speakerPartitioning: true` and pass that count. If the answer is yes but the count is unknown, set `speakerPartitioning: true` and omit `speakersCount` for automatic detection.\n- Skip the speaker question when the user already specified whether speaker labelling is wanted. Never guess the number of speakers or attempt to determine it before transcription.\n- Set `channelIdentification: true` only when the user requests channel identification. Speaker diarization and channel identification may both be enabled when requested; omit `speakersCount` in that combination so AssemblyAI detects speakers automatically within each channel.\n- Use the tool defaults for other unspecified options. Pass a language, translation target, transcription prompt, PII redaction, or profanity filtering only when the user requests it or supplies the value.\n- Do not claim that advanced options unavailable in `upload_and_transcribe` were enabled.\n- Do not automatically retry an upload. A retry can create a duplicate Traspion file.\n\nTreat the tool result as authoritative:\n\n- If transcription completed, retrieve any remaining transcript pages needed for the user's request. Present the text in ChatGPT's native editable view when available; otherwise show it in the conversation. Never claim that a partial page is the complete transcript.\n- If validation or transcription is still processing, preserve the returned `fileId`, explain the status and suggested follow-up, and do not claim completion.\n- On an explicit request to check progress, call `get_transcription` with that `fileId`. If it is still processing, report the status and stop; do not repeatedly poll within the same response or upload the file again.\n- If the account lacks transcription balance, report the failure and present the exact Traspion billing-page link returned by the tool. Do not retry the upload or rewrite the link.\n"
}