voice-notes
Skill9 starsChanged 4 days agoMITView on GitHub
Handle Slack voice notes and other audio messages: hear the clip natively when the model can, otherwise transcribe it through the online transcription backend.
We split this skill into the rules it gives an agent
6 rules. About 0 are common, 3 are distinctive, and 3 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈3
Distinctive · estimate
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
- Very specific: line 8
Read audio messages from their spoken content.
- Distinctive: line 20
Run this command to transcribe one file:
- Distinctive: line 38
When the command fails, report that to the user and do not guess the words.
- Very specific: line 44
When Slack hosts the file, fetch it through url_private_download with the app bearer.
- Very specific: line 47
Use the online transcription result.
- Distinctive: line 50
Before you quote voice content outside the thread, redact it.
≈ “Common” is an estimate: at least 1 in 5 of the 308,232 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
Voice Message Handling
Read audio messages from their spoken content. A Slack voice note arrives as
an empty-text message carrying one hosted audio/m4a file. Route in this
order:
1. Native audio input (preferred)
A conversation model that declares audio input hears the clip directly: wording and prosody reach the reasoning turn in one request. Backend lineups change, so take the deployment's declared modality over memory.
2. Online transcription
Run this command to transcribe one file:
charliebot voice transcribe <file>Purpose: the transcript comes from the same backend that the voice upload uses.
The command reads these inputs:
- WAV, MP3, and MP4 with AAC;
- Ogg with Vorbis or Opus, and WebM with Opus;
- raw 16 kHz mono s16 PCM (
.pcm,.s16).
Purpose: a model converts a file first when its container is not on this list.
The command sends the whole clip through the online transcription backend. It uses
voice.default_backend from the profile's config. The backend's key comes from
credentials.yaml. The transcript goes to stdout.
When the command fails, report that to the user and do not guess the words.
Purpose: a guessed transcript answers words the user never said.
Slack specifics
When Slack hosts the file, fetch it through url_private_download with the app bearer.
Purpose: the command reads a local file, so the clip must be on disk first.
Use the online transcription result.
Purpose: Slack's files.info transcript can take minutes.
Before you quote voice content outside the thread, redact it. Purpose: voice content is PII.
In this repository21 other files it gives its agents.
- AGENTS.mdAGENTS.md
- charliebotSkill
- claude-versionSkill
- cli-tool-trapsSkill
- feishuSkill
- file-serverSkill
- gmailSkill
- google-docsSkill
- google-oauthSkill
- google-sheetsSkill
- improve-goalSkill
- improve-workerSkill
- language-portSkill
- linearSkill
- llm-context-guidelineSkill
- perfettoSkill
- plan-approvalSkill
- pytorch-gpu-trainingSkill
- skill-managementSkill
- slackSkill
- writing-styleSkill
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
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.