Vocal Chat

PassAudited by ClawScan on May 1, 2026.

Overview

The skill is coherent for WhatsApp voice chat, but users should verify the undeclared local tools and understand it can send chat replies on their behalf.

Install this only if you want the agent to handle WhatsApp voice replies. Before using it, make sure the referenced local speech tools are trusted and available, and enable the mode only in conversations where automatic text and audio responses are appropriate.

Findings (3)

Artifact-based informational review of SKILL.md, metadata, install specs, static scan signals, and capability signals. ClawScan does not execute the skill or run runtime probes.

What this means

The agent may send text and voice-note responses to WhatsApp conversations when the skill is triggered.

Why it was flagged

The skill delegates the agent to send messages/files through the user's WhatsApp chat channel. This is expected for the purpose, but it is still account/messaging authority the user should notice.

Skill content
Send the resulting `.ogg` file back to the user as a voice note. ... send `/tmp/reply.ogg` via `message` tool with `filePath`.
Recommendation

Use this only in chats where automated replies are acceptable, and keep control over when walkie-talkie mode is enabled.

What this means

If the referenced local tools are missing, replaced, or untrusted, transcription or TTS behavior may not match the skill description.

Why it was flagged

The instructions depend on local helper scripts and binaries that are not included with this instruction-only artifact and are not declared in the registry requirements. This is purpose-aligned but leaves setup/provenance to the user.

Skill content
Use `tools/transcribe_voice.sh` ... generate speech using `bin/sherpa-onnx-tts`. ... Use local tools only (ffmpeg, whisper-cpp, sherpa-onnx-tts).
Recommendation

Install and use only trusted copies of ffmpeg, whisper-cpp, sherpa-onnx-tts, and any helper scripts, and verify their paths before use.

What this means

The agent may run a local TTS binary and write an audio file to `/tmp/reply.ogg` as part of responding.

Why it was flagged

The skill documents local command execution to generate TTS audio. This is central to the voice-chat purpose and is not hidden, but it is still local execution the user should understand.

Skill content
bin/sherpa-onnx-tts /tmp/reply.ogg "Tu mensaje aquí"
Recommendation

Use trusted local binaries and avoid enabling the skill in environments where local command execution is not desired.