Skill flagged — suspicious patterns detected
ClawHub Security flagged this skill as suspicious. Review the scan results before using.
Voice Wake Say
v1.0.1Speak responses aloud on macOS using the built-in `say` command when user input indicates Voice Wake/voice recognition (for example, messages starting with "User talked via voice recognition on <device>").
⭐ 5· 8.1k·56 current·58 all-time
MIT-0
Download zip
LicenseMIT-0 · Free to use, modify, and redistribute. No attribution required.
Security Scan
OpenClaw
Suspicious
medium confidencePurpose & Capability
The skill explicitly depends on the macOS 'say' command, but the registry metadata lists no OS restriction and no required binaries. That's an incoherence: someone installing this may expect cross-platform behavior, yet the instructions only work on macOS.
Instruction Scope
Instructions are narrowly scoped (trigger only when the message starts with the exact phrase) and limit use to local 'say' (no cloud TTS). However the runtime snippet references a runtime variable ($SPOKEN_TEXT) and optional env vars (SAY_VOICE, SAY_RATE) that are not declared in metadata. The SKILL.md instructs running shell commands (printf | say) which is expected but will execute local commands.
Install Mechanism
Instruction-only skill with no install spec or code files — lowest-risk installation model. Nothing is written to disk by the skill itself.
Credentials
No required credentials or config paths are requested (good). The instructions mention optional env vars for voice and rate; these are reasonable but not declared in requires.env — this is a minor mismatch to clarify.
Persistence & Privilege
always:false and default invocation settings are appropriate. The skill does not request persistent system-wide changes or extra privileges.
What to consider before installing
This skill is simple and appears to do what it says (run the local macOS 'say' command when a message starts with the exact trigger phrase). Before installing: (1) confirm your agent runs on macOS and that /usr/bin/say (or equivalent) exists — the registry metadata should explicitly state macOS-only; (2) confirm you are comfortable with the agent running shell commands locally (it will execute printf | say); (3) ask the author to declare optional env vars (SAY_VOICE, SAY_RATE) and the expected runtime variable ($SPOKEN_TEXT) in the metadata for clarity; (4) verify you trust the agent not to route spoken text to external services (the README deliberately avoids cloud TTS). These mismatches look like sloppy metadata rather than malicious intent, but fix the OS/binary declarations before relying on the skill.Like a lobster shell, security has layers — review code before you run it.
latestvk97d6v1v8nkfdzt5hdjd81n1g57zwzyb
License
MIT-0
Free to use, modify, and redistribute. No attribution required.