voice-note-ingest

Ingest voice notes into a knowledge base with verbatim transcript preservation and directory routing.

Updated Aug 17, 2026
One-click install
npx skills add https://github.com/AMC-JTC/gbrain-1 --skill voice-note-ingest-amc-jtc
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: voice-note-ingest
Source: https://github.com/AMC-JTC/gbrain-1/tree/main/plugin/skills/voice-note-ingest
Command: npx skills add https://github.com/AMC-JTC/gbrain-1 --skill voice-note-ingest-amc-jtc

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? Voice notes and audio memos often contain raw, unpolished ideas that get lost or diluted when paraphrased into notes. This Skill captures the user's exact words, stores the original audio, and files the content into the correct knowledge-base directory so nothing is lost or misfiled. ## Core Features & Use Cases - Verbatim Transcript Preservation: Stores the exact transcript in a block-quoted "User's Words" section, never paraphrasing or cleaning up filler words, with a link back to the original audio file. - Decision-Tree Routing: Routes content to originals/, concepts/, people/, companies/, ideas/, personal/, or voice-notes/ based on what the note is about, with cross-linking when multiple categories apply. - Entity Cross-Linking: Adds timeline back-links from every mentioned person or company page to the new voice-note page. - Use Case: You send a voice memo from Telegram with a spontaneous product idea mentioning a colleague. The Skill stores the audio, transcribes it verbatim, creates an ideas/ page with your exact words, and updates the colleague's people/ page with a back-link. ## Quick Start Ingest this voice note I just sent, transcribe it verbatim, and file it into the right brain directory with the original audio attached.

Frequently Asked Questions about voice-note-ingest

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I ingest a voice memo into my knowledge base?

Send the audio through any channel such as Telegram or a voice memo upload, and the Skill stores the original audio, transcribes it verbatim, routes the content to the correct directory, and creates a brain page with your exact words preserved in a block quote.

How does voice note routing decide where to file content?

A decision tree is applied in order: original ideas go to originals/, world concepts to concepts/, person updates to people/, company updates to companies/, product ideas to ideas/, personal reflections to personal/, and anything else to voice-notes/ as a catch-all.

What transcription tools work for audio notes?

The Skill uses the transcript provided by the host agent when available. Otherwise it transcribes with Groq Whisper or OpenAI Whisper, segmenting audio files larger than 25MB with ffmpeg first.

Can one voice note create pages in multiple directories?

Yes. When a note covers multiple categories, the Skill creates the primary page in the first matching directory and cross-links to the others, such as creating an originals/ page while also updating a mentioned person's timeline.

Why does the Skill refuse to paraphrase transcripts?

The exact unpolished phrasing is treated as the primary signal, so the transcript section is preserved verbatim including hesitations and filler words. Only the separate Analysis section contains the agent's interpretation.