stt

Transcribe audio files using SkillBoss API Hub STT

Security Scan

The skill is coherent: it reads a local audio file, base64-encodes it, and sends it to a single third‑party STT endpoint using a SKILLBOSS_API_KEY — this matches the stated purpose.

Purpose & Capability

Name/description, SKILL.md, and the included transcribe.py all describe sending audio to SkillBoss API Hub STT and require an API key. The single required credential (SKILLBOSS_API_KEY) is appropriate for this purpose. Minor naming inconsistencies (skill slug 'modesty-stt' and SKILL.md name 'gemini-stt') are cosmetic but not security-relevant.

Instruction Scope

Runtime instructions and the script only read the provided audio file, base64-encode it, and POST JSON to the external API. They do not attempt to read unrelated files, other environment variables, or system configuration. The script does transmit the full audio contents to the external service, which is consistent with its stated function.

Install Mechanism

No install spec; the skill is instruction/code-only and uses only the Python stdlib. Nothing is downloaded or extracted during install, so there is no elevated install-time risk.

Credentials

Only SKILLBOSS_API_KEY is required and used by the script. No unrelated credentials, config paths, or excessive environment access are requested.

Persistence & Privilege

always is false and the skill does not request persistent or system-wide privileges. It does not modify other skills or system configuration.

Assessment

This skill will upload the entire audio file (base64) to api.heybossai.com using SKILLBOSS_API_KEY. Before installing, confirm you trust the SkillBoss/HeyBossAI service and its privacy policy, and avoid transcribing sensitive audio unless you're comfortable with third‑party processing. Manage the API key carefully (use least-privilege/rotating keys if possible) and restrict where you run the skill to systems allowed to make external network requests. Note the minor naming inconsistencies and lack of a homepage — if you need organizational provenance, ask the publisher for more information.

Speech-to-Text Skill

Transcribe audio files using SkillBoss API Hub. Automatically routes to the best available STT model.

Authentication

Set SKILLBOSS_API_KEY in environment (e.g., ~/.env or ~/.clawdbot/.env)

Requirements

• Python 3.10+ (no external dependencies)
• SKILLBOSS_API_KEY environment variable

requires.env

SKILLBOSS_API_KEY

Supported Formats

• .ogg / .opus (Telegram voice messages)
• .mp3
• .wav
• .m4a

Usage

# Transcribe an audio file
python ~/.claude/skills/gemini-stt/transcribe.py /path/to/audio.ogg

# With Clawdbot media
python ~/.claude/skills/gemini-stt/transcribe.py ~/.clawdbot/media/inbound/voice-message.ogg

Options

Option	Description
<audio_file>	Path to the audio file (required)

How It Works

1. Reads the audio file and base64 encodes it
2. Authenticates using SKILLBOSS_API_KEY
3. Sends to SkillBoss API Hub /v1/pilot with type: stt
4. SkillBoss API Hub automatically routes to the best available STT model
5. Returns the transcribed text from data.result.text

Example Integration

For Clawdbot voice message handling:

# Transcribe incoming voice message
TRANSCRIPT=$(python ~/.claude/skills/gemini-stt/transcribe.py "$AUDIO_PATH")
echo "User said: $TRANSCRIPT"

Error Handling

The script exits with code 1 and prints to stderr on:

• No SKILLBOSS_API_KEY set
• File not found
• API errors

Notes

• SkillBoss API Hub automatically selects the best STT model
• No external Python dependencies (uses stdlib only)
• Automatically detects MIME type from file extension