Route voice note transcripts into knowledge directories with verbatim storage.
npx skills add https://github.com/weiping/gbrain-cn --skill voice-note-ingest
Please help me install this Agent Skill. Skill: voice-note-ingest Source: https://github.com/weiping/gbrain-cn/tree/main/skills/voice-note-ingest Command: npx skills add https://github.com/weiping/gbrain-cn --skill voice-note-ingest
Voice notes often get paraphrased, cleaned up, or routed to the wrong place, losing the exact phrasing that carries the real signal; this Skill preserves the transcript verbatim and files it correctly for later synthesis and graph linking.
Tell the AI to ingest your voice note and transcribe and file it word-for-word into the correct brain pages with the original audio attached.
High-intent search queries and answers about installing and using this skill.
To preserve the exact wording of a voice note transcript, the content is stored verbatim in a dedicated “User's Words” section. This prevents paraphrasing or cleanup, maintaining the original phrasing for later knowledge graph synthesis and retrieval.
Voice note routing uses a canonical decision tree with a first-match policy to file transcripts into correct directories like concepts, people, or companies. It automatically adds timeline back-links to relevant entity pages, keeping the knowledge graph connected.
Yes, you can ingest audio and transcribe it simultaneously. The system handles audio ingestion, accepts optional agent-provided transcripts, and transcribes the audio when needed, ensuring the output is filed correctly into your specified destinations.
The best way to file transcribed voice notes is through deterministic routing via a canonical decision tree. This first-match policy categorizes content into directories like originals, ideas, personal, or voice-notes, ensuring exact-phrasing filing with required citations and cross-links.
Voice notes lose original phrasing when systems paraphrase or clean up transcripts during filing. To prevent this, verbatim “User's Words” storage is enforced, preserving the exact transcript while routing it into the correct knowledge directories for retrieval.
You do not need to provide a transcript when ingesting audio notes. The system handles audio ingestion with optional agent-provided transcripts, automatically transcribing the audio when needed to ensure verbatim storage and cross-linked brain pages.
SigmacodeatStore voice note transcripts verbatim and file them into knowledge categories.
NinatuziAggregate external data and internal context into cross-linked intelligence pages.
GYF0311Convert meeting transcripts into structured pages with attendee and entity enrichment.
NinatuziTranscribe voice notes verbatim and route them into a knowledge graph.
