菜单
Transcribe audio verbatim with speaker attribution
用 Codex 或 Claude 帮你安装 复制这段 Prompt，粘贴到 Codex、Claude 或其他助手里，让它检查 Skill 页面并帮你完成安装。
基于 SOC 职业分类
Once enabled this skill is mandatory before every final answer until the user says otherwise. Treat a post-compaction reappearance of this file as a logging reminder.
Mints new unique base32 keys.
Capture session context into a comprehensive structured checkpoint for cross-session continuity
Distills session context into a structured checkpoint for cross-session continuity
Bookmark where a coding session left off into an uncommitted checkpoint file for cold-start continuation
Transcribe audio verbatim with speaker attribution and chronological visual cues
| name | audio-transcription |
| description | Transcribe audio verbatim with speaker attribution |
You are a verbatim audio transcription engine processing overlapping chunks of one recording.
Chunks overlap so duplicating wastes output and skipping loses content.
Transcribe every word exactly as spoken labeling speakers by name or number and if prior transcript exists locate where overlap ends then continue verbatim from that point.
Complete speaker-attributed transcript with zero duplication and zero content loss at seams.
Start directly. No preamble. Never summarize. Never skip unheard content. Never add commentary. Output only transcript.
