Walkie-Talkie Mode
PassAudited by ClawScan on May 1, 2026.
Overview
This is a coherent WhatsApp voice-reply skill, but users should verify the local audio tools it calls and be comfortable with automatic WhatsApp replies.
Before installing, make sure you trust the local transcription and TTS tools referenced by the skill, and only use it in WhatsApp contexts where automatic text-plus-audio replies are acceptable.
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.
The skill may not work unless those tools already exist, and any locally installed helper code would need to be trusted separately.
The skill depends on local helper paths that are not included in the provided file manifest or install spec. Because these helpers are central to the disclosed purpose, this is a setup/provenance note rather than a suspicious mismatch.
Use `tools/transcribe_voice.sh` to get the text. ... generate speech using `bin/sherpa-onnx-tts`.
Before use, verify that the referenced transcription and TTS tools are present, trusted, and located at the expected paths.
The assistant may invoke local audio-generation commands as part of the voice reply workflow.
The skill documents running a local executable to generate speech. This is expected for a local TTS workflow, but it is still local command execution that users should recognize.
bin/sherpa-onnx-tts /tmp/reply.ogg "Tu mensaje aquí"
Use only trusted local TTS binaries and ensure the command path cannot be replaced by an untrusted executable.
When active, the skill can send text and audio replies through the user's WhatsApp messaging channel.
The skill uses the assistant's messaging capability to send files back over WhatsApp. This is core to the skill's purpose, but it does rely on delegated messaging authority.
Send the resulting `.ogg` file back to the user as a voice note. ... send `/tmp/reply.ogg` via `message` tool with `filePath`.
Enable it only for conversations where automatic voice and text replies are acceptable, and review any platform-level permissions granted to the assistant.