中文 OpenClaw 插件
PassAudited by ClawScan on May 10, 2026.
Overview
This instruction-only skill is coherent, but it teaches creating persistent OpenClaw hooks that can change prompt context, so users should review any plugin before enabling it.
Install this only if you intend to build OpenClaw plugins or hooks. Review any generated hook code before enabling it, because enabled hooks can persist and change the prompts or messages the agent uses.
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.
A plugin created from this guidance could change how the agent interprets future prompts.
The before_prompt_build hook example can change the system message or user message before the agent sees it.
context.systemMessage = "你的系统消息"; ... // context.messages[0].content = "修改后的内容";
Only create or enable hooks whose behavior you understand, and review prompt/message modifications before using them.
Hook code can run during OpenClaw prompt handling and affect the agent runtime.
The skill instructs users to create JavaScript hook files that OpenClaw will execute as part of plugin behavior.
hooks/before_prompt_build.js ... module.exports = async function(context) { ... return context; };Review hook JavaScript carefully, avoid untrusted code, and keep hook behavior narrow and documented.
A plugin may continue affecting future sessions after the initial task.
The documented plugin location and enable command create persistent OpenClaw behavior until the plugin is disabled.
~/.openclaw/plugins/<plugin-name>/ ... openclaw plugin enable <plugin-name> ... openclaw plugin disable <plugin-name>
Use clear plugin names, list enabled plugins periodically, and disable or remove plugins that are no longer needed.