wechat-mp-writer-skill-mxx
v1.0.0微信公众号文章全流程写作助手。支持热点选题、文章撰写、AI去味润色（去除机器感）、配图建议、一键发布到公众号草稿箱。适用于技术博主、自媒体创作者、企业公众号运营等场景。当用户需要写公众号文章、润色内容、寻找配图或发布到微信时使用此skill。
⭐ 32· 4k·30 current·30 all-time
MIT-0
Download zip
LicenseMIT-0 · Free to use, modify, and redistribute. No attribution required.
Security Scan
OpenClaw
Benign
high confidencePurpose & Capability
The name/description (WeChat MP writer + publish to draft) match the instructions. The skill does not request WeChat credentials or other secrets; instead it delegates publishing to a separate wechat-mp-publisher skill if available. This is coherent, but users should understand the skill cannot itself authenticate to WeChat — one-click publishing requires the external publisher skill (and that skill's credentials).
Instruction Scope
SKILL.md contains purely writing, style, image-suggestion, and publishing-or-delegate instructions. It does not tell the agent to read local files, environment variables, or any unrelated system paths, nor does it embed hidden endpoints. It mentions 'search latest hotspots' and 'search for images' without specifying external services, which is reasonable for an instruction-only skill but leaves implementation of those searches up to the agent or other tools.
Install Mechanism
No install spec and no code files are present (instruction-only), so nothing will be written to disk or downloaded during install — lowest-risk install profile.
Credentials
The skill declares no required environment variables, credentials, or config paths. That is proportionate to an instruction-only writing assistant that delegates actual publishing to another tool.
Persistence & Privilege
always is false and the skill does not request system-wide persistence. The skill can be invoked autonomously by agents (platform default), which is expected for skills; this alone is not a security concern here.
Assessment
This skill is an instruction-only WeChat article assistant and appears internally consistent. Before installing or using it: (1) know that publishing to WeChat is delegated — you must install/configure a separate wechat-mp-publisher skill (and that other skill will need your WeChat credentials); (2) review the publisher skill’s credential handling and permissions before granting access; (3) be aware the 'AI 去味' guidance encourages adding personal anecdotes — avoid inventing real people/events or false claims; (4) if you want to limit automatic behavior, restrict autonomous agent invocation or require manual confirmation before publishing; and (5) check image sourcing/licensing when following the skill’s image suggestions.Like a lobster shell, security has layers — review code before you run it.
latestvk975z4btrr3xcv8wbnabcjzfan81zrtr
License
MIT-0
Free to use, modify, and redistribute. No attribution required.