voice-chat-mode
v0.1.6在用户明确要求中文语音聊天或中文语音模式时激活。
⭐ 3· 72·0 current·0 all-time
bysongkuakua@liang9886703
MIT-0
Download zip
LicenseMIT-0 · Free to use, modify, and redistribute. No attribution required.
Security Scan
OpenClaw
Benign
high confidencePurpose & Capability
Name, description, and the SKILL.md all align: the skill's sole purpose is to rewrite replies into natural spoken Chinese suitable for TTS. There are no unexpected binaries, env vars, or installs required.
Instruction Scope
Instructions are narrowly focused on producing spoken Chinese and include many style rules. One notable constraint is the requirement that final output contain only the spoken body text (no explanations, stage directions, or meta-comments). That is coherent for TTS but could suppress assistant-side disclaimers/refusals or meta-safety messages if the platform or implementer expects those to be included. This is a design choice rather than evidence of malice, but you should confirm how the host agent enforces policy-level refusals when a skill demands 'only spoken text.'
Install Mechanism
Instruction-only skill with no install spec and no code files. Lowest-risk install profile: nothing is written to disk or downloaded.
Credentials
No environment variables, credentials, or config paths are requested — proportionate to the stated purpose.
Persistence & Privilege
Skill requests that once enabled it remain active until the user turns it off. 'always' is false, so it won't be auto-included by the system, but if the agent or user enables it, expect it to keep rewriting replies in voice mode until disabled. Combined with autonomous invocation (normal default), this can change the agent's behavior across multiple turns; confirm how enabling/disabling is presented to users.
Assessment
This skill appears coherent and low-risk: it only provides formatting/instruction rules for spoken Chinese and asks for no secrets or installs. Before enabling, check two things: (1) How your agent/platform enforces safety refusals or content warnings — the skill asks the assistant to emit only spoken text (no meta-comments), so confirm that required refusals or safety notices won't be inadvertently suppressed; (2) How to disable the skill at runtime — the SKILL.md requests persistent activation until explicitly turned off, so make sure you can easily toggle it if it starts affecting responses you don't want in voice-style. If those two points are acceptable, the skill is reasonable to install. Test with harmless prompts first.Like a lobster shell, security has layers — review code before you run it.
latestvk97fhk63gqr70kk6xbw0x9js3s84qc76
License
MIT-0
Free to use, modify, and redistribute. No attribution required.