Generate high-quality English (and multilingual) audio using Microsoft Edge TTS. Use when the user asks to "speak this", "pronounce", "read aloud", "say this...

This skill appears to do what it says: call the edge-tts client to produce MP3 audio and return it. Before installing, verify that you are willing to (a) let the agent run a local binary at /root/.local/bin/edge-tts (or change the path to where you installed it), and (b) allow edge-tts to communicate with Microsoft services — any text you send for TTS may be transmitted to Microsoft endpoints by that client. Because the registry entry includes no install step, consider installing edge-tts yourself (pipx) and confirming the binary path. If you need stricter guarantees, ask the author to add an explicit install spec, or inspect/replace the script to run a vetted TTS binary. Finally, ensure sensitive text is not sent to TTS if you do not want it shared with external services.

Type: OpenClaw Skill Name: edge-tts-english Version: 1.0.2 The `scripts/speak.sh` file is vulnerable to shell injection. The `$TEXT`, `$VOICE`, and `$OUTPUT` arguments are directly interpolated into the `edge-tts` command without explicit sanitization. If the OpenClaw agent passes unsanitized user input to this script, an attacker could inject arbitrary shell commands (e.g., `'; rm -rf / #'`) leading to remote code execution. This is a critical vulnerability, but not evidence of intentional malicious design within the skill itself.