Repository inventory

ntdrun/ari-skills

Skills indexed from this repository, with install-style signals scoped to the repo.
2 skills0 GitHub stars0 weekly installsShellGitHubOwner profile

Overview

This skill captures and preserves every idea from meetings, negotiations, brainstorms, and business conversations. It converts uploaded audio to a raw verbatim transcript or accepts an existing transcript, then structures a comprehensive meeting document focused on completeness rather than polish. The output is a Markdown protocol with decisions, ideas, questions, and tasks preserved verbatim where needed.

How this skill works

When a user uploads audio (mp3, wav, m4a, etc.) the skill transcribes the file fully and saves the raw transcript for verification. If a transcript is already provided, it accepts it unchanged as the source. It analyzes participant count, meeting purpose, and content types, then generates a Markdown file with standardized sections: key artifacts, open questions, tasks, and discussion flow. Finally it checks the transcript to ensure every meaningful unit of thought is captured and reports counts of artifacts, questions, and tasks.

When to use it

  • Any time the user uploads an audio recording of a meeting, negotiation, call, brainstorming session, or business conversation.
  • When the user pastes or uploads a raw transcript and asks to process or make a protocol.
  • When the user mentions meeting-related keywords (meeting, call, negotiation, protocol, transcription, minutes).
  • When the user says something like “here’s the recording—process it” or “make a protocol from this.”

Best practices

  • Keep the raw transcript untouched until analysis; preserve vocal markers, names, and numbers verbatim.
  • Prioritize completeness: include imperfect formulations rather than omitting uncertain thoughts.
  • Separate remarks by speaker when multiple participants are present; if speaker ID is unclear, use contextual cues.
  • Label every significant thought as an artifact type (Decision, Strategy, Principle, Idea, Observation).
  • Run a final completeness pass comparing the document to the saved transcript and add any missed items.

Example use cases

  • Turn a client negotiation audio into a protocol with explicit decisions and follow-up tasks.
  • Convert a brainstorming session recording into a list of ranked ideas and responsible owners.
  • Process a remote team standup call to extract open questions and action items for tracking.
  • Accept a pasted transcript from a workshop and produce a downloadable Markdown meeting record with artifacts and chronology.

FAQ

No — verbatim wording is preserved in the raw transcript. Editorial cleanup happens only to remove filler words and to structure thoughts into artifact blocks while retaining original meaning and important phrasings.

What output formats are provided?

The primary output is a ready-to-download Markdown (.md) meeting document with sections for artifacts, open questions, tasks, and discussion flow. A brief chat summary with artifact/question/task counts is also provided.

2 skills

More from this maintainer
Other repositories and skills published under the same GitHub owner.
Skills library
Jump back to the full directory or explore grouped topics.
Built by
VeilStrat
AI signals for GTM teams
© 2026 VeilStrat. All rights reserved.All systems operational