Case 01177

Skill knownasnaffy/prompthound/dataset/case_01177

A fast, offline static risk analysis CLI for AI agent skill files. Detects malicious instructions, steganographic payloads, and dangerous capability chains.

Install

npx -y skills add knownasnaffy/prompthound --skill case_01177

Assembled from the repository path, not quoted from the project. Check it against their README if it does not work.

2 things to look at

no licenseNo license file was found in the repository. Code published without one is not open source by default, so using it at work is a question for whoever answers licensing questions where you are.
0 stars0 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.

What its author says it does

Copied from the file, not written here

Transcribe audio files (voice notes, recordings, podcasts) to text via the Speechmatics batch transcription API. Use when the user asks to transcribe audio, convert speech to text, or get a transcript of a recording, and Speechmatics is configured.

SKILL.md

3.3 KB, 644 tokens by cl100k_base, as published. Nobody here has run it

Cleanup Review

python scripts/_purge_cache.py is an optional cleanup helper for the Speechmatics (batch transcription) skill. It may recursively remove /tmp/skill_build/*, /tmp/skill_build/.cache. Show the exact paths first, explain whether they are skill-private temporary paths or user state, and run the cleanup only after the user confirms.

Speechmatics (batch transcription)

Transcribe an audio file via Speechmatics' async batch API. Submits a job, polls until complete, then writes the transcript.

Quick start

{baseDir}/scripts/transcribe.sh /path/to/audio.m4a

Defaults:

Language: en
Operating point: enhanced (better accuracy; use standard for faster/cheaper)
Output: <input>.txt in the same directory
Poll interval: 3s, timeout: 600s

Useful flags

{baseDir}/scripts/transcribe.sh /path/to/audio.ogg --language da
{baseDir}/scripts/transcribe.sh /path/to/meeting.wav --operating-point standard
{baseDir}/scripts/transcribe.sh /path/to/call.mp3 --diarization speaker
{baseDir}/scripts/transcribe.sh /path/to/audio.m4a --format json --out /tmp/transcript.json
{baseDir}/scripts/transcribe.sh /path/to/audio.m4a --format srt --out /tmp/subs.srt
{baseDir}/scripts/transcribe.sh /path/to/long.wav --timeout 1800

Formats: txt (default, plain text), json (Speechmatics json-v2 with word timings), srt (subtitles).

API key

The script reads the API key from (in order):

--api-key <key> flag
SPEECHMATICS_API_KEY environment variable (set by openclaw from the entry below)
skills.entries.speechmatics.apiKey in $OPENCLAW_CONFIG_PATH (default ~/.openclaw/openclaw.json)

Configure via openclaw.json:

{
  skills: {
    entries: {
      speechmatics: {
        apiKey: "SPEECHMATICS_KEY_HERE",
      },
    },
  },
}

Override the API base (e.g. EU region or a proxy) with --base-url or SPEECHMATICS_BASE_URL. Default: https://asr.api.speechmatics.com/v2.

Notes

Supports most common audio formats (wav, mp3, m4a, ogg, flac, mp4, etc.) — Speechmatics transcodes server-side.
File size limit: 2 GB per job.
Batch jobs complete in roughly 1:10 wallclock to audio duration on enhanced; standard is faster.
Always confirm any destructive follow-up (e.g. replying based on a transcript) before acting.

What ships with it: 2 files

4.3 KB alongside SKILL.md, 2 of them executable

scripts/

_purge_cache.pyruns499 B
transcribe.shruns3.8 KB