Speechmatics

Transcribe audio files (voice notes, recordings, podcasts) to text via the Speechmatics batch transcription API. Use when the user asks to transcribe audio,...

Security Scan

The skill's code, declared requirements, and instructions are consistent with a simple Speechmatics batch-transcription helper; it only needs an API key plus curl/jq and uploads the provided audio to Speechmatics for transcription.

Purpose & Capability

Name/description match what the files do: submit an audio file to Speechmatics' batch API, poll for completion, and save a transcript. Required binaries (curl, jq) and SPEECHMATICS_API_KEY are appropriate and expected for this purpose.

Instruction Scope

The runtime script sticks to the transcription workflow (reading an audio file, POSTing it, polling, saving output). It also optionally reads an OpenClaw config file ($OPENCLAW_CONFIG_PATH or ~/.openclaw/openclaw.json) to obtain an API key and accepts a SPEECHMATICS_BASE_URL override; these are reasonable fallbacks but the SKILL.md/manifest do not list that config path explicitly, so users should be aware the script may read that file for credentials.

Install Mechanism

Install spec only references installing jq (brew formula) and apt packages for curl/jq — standard packages from normal package managers. No downloads from arbitrary URLs or extracted archives are used.

Credentials

The only required secret is SPEECHMATICS_API_KEY (declared as primary). The script also supports an optional SPEECHMATICS_BASE_URL and will read OPENCLAW_CONFIG_PATH for a stored key if present; those optional accesses are reasonable but the manifest didn't declare the config path or the optional base-url env var explicitly.

Persistence & Privilege

The skill is not always-enabled and does not request elevated or persistent system privileges, nor does it modify other skills or global configs beyond reading an optional user config for an API key.

Assessment

This skill appears to do exactly what it claims: upload a provided audio file to Speechmatics and save the returned transcript. Before installing, consider: (1) it will send your audio to Speechmatics — don't transcribe sensitive recordings unless you trust the service and its region/retention policies; (2) the script will look for an API key in SPEECHMATICS_API_KEY, an --api-key flag, or in your OpenClaw config (~/.openclaw/openclaw.json) — if you don't want it to read that file, remove the key there or set the env var explicitly; (3) the script accepts SPEECHMATICS_BASE_URL (useful for private/regional endpoints or proxies) — verify this isn't pointed at an unexpected proxy; (4) it requires curl and jq (installable via apt/brew as documented). These are normal for this skill; no suspicious network endpoints or obfuscated code were found.

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):

1. --api-key <key> flag
2. SPEECHMATICS_API_KEY environment variable (set by openclaw from the entry below)
3. 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.