agentleFS
Sign inStart free
QuixiAIQuixiAI/Hexis/skills/installed/audio-analysis/SKILL.md

audio-analysis

Skill623 starsChanged 10 months agoMITView on GitHub

Analyze a local recording for speakers and label timestamped transcript segments without uploading audio

6 other projects use a version of this.

Adopted by

Most recently taken up by

The file

We split this skill into the rules it gives an agent

6 rules. 0 are common, 3 are distinctive, and 3 are very specific to this project.

  • 0

    Common

    Most of the 4 edited versions we read keep these.

  • 3

    Distinctive

    This author's own choices. Worth a look.

  • 3

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Use device-local speaker diarization when the user asks who spoke when in a recording.

    line 14
  2. Very specific:

    Call analyze_local_audio with action=start.

    line 22
  3. Distinctive:

    If the job fails, report its cause and exact next step from the status.

    line 26
  4. Distinctive:

    Do not enable autonomous audio analysis without the user's explicit choice.

    line 30
  5. Very specific:

    Do not copy generated artifacts into a source repository.

    line 31
  6. Very specific:

    Only request emotion_heuristics when the user asks for them.

    line 32

“Common” is measured: kept by at least 6 in 10 of the 4 edited versions of this file we could read. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Local Audio Analysis

Use device-local speaker diarization when the user asks who spoke when in a recording.

For a plain speech-to-text request, use transcribe; it follows the local/cloud choice in Settings and does not require a diarization job.

Method

  1. Confirm the exact local audio path and, when available, a Whisper JSON path with timestamped segments.
  2. Call analyze_local_audio with action=start. The tool requires operator approval and never uploads audio.
  3. Report the returned cache directory and immediately explain that analysis continues in the background.
  4. Poll with action=status and the same audio_path until the job completes or fails.
  5. On completion, use the returned diarized JSON or SRT artifact. Preserve SPEAKER_* labels unless the user supplies identities.
  6. If the job fails, report its cause and exact next step from the status. Never present acoustic heuristics as a reliable reading of emotion.

Boundaries

  • Do not enable autonomous audio analysis without the user's explicit choice.
  • Do not copy generated artifacts into a source repository; they belong in the Hexis cache.
  • Only request emotion_heuristics when the user asks for them. Label the result as a coarse local acoustic estimate.

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

Check your setup, free