技术类文章生成
v1.0.0微信公众号技术文章端到端生成流程。当用户给出一个技术主题，希望生成一篇完整的微信公众号文章时使用此技能。流程包含四步：（1）资料收集与主题分析调研；（2）根据大纲撰写正文；（3）质量检查（目标8.5/10分）；（4）错误修复与润色优化。用户需提供主题、大纲（可选）和目标字数。
⭐ 1· 53·0 current·0 all-time
by澄@sariel20
MIT-0
Download zip
LicenseMIT-0 · Free to use, modify, and redistribute. No attribution required.
Security Scan
OpenClaw
Benign
high confidencePurpose & Capability
Name/description (WeChat technical article generation) align with the SKILL.md: the instructions describe research, drafting, review, and polishing steps. No unexpected binaries, env vars, or installs are requested.
Instruction Scope
The runtime instructions tell the agent to collect external documentation, blogs, and code examples and to mark sources — this matches a research step but implies web access and potential copyright/licensing considerations. The SKILL.md does not direct the agent to read system files, environment variables, or send data to arbitrary endpoints.
Install Mechanism
No install spec or code files are present (instruction-only), so nothing is written to disk or fetched during installation.
Credentials
The skill requires no environment variables, credentials, or config paths — requested access is proportional to the described task.
Persistence & Privilege
always is false and the skill has no install hooks or modifiers that alter other skills or system settings. It does not request elevated or persistent privileges.
Assessment
This skill is instruction-only and internally consistent, so installing it poses minimal system-level risk. Before using it: (1) avoid pasting proprietary or sensitive code into the prompt because the workflow encourages external research and reuse of examples; (2) verify and run any generated code yourself — the skill requires 'runnable' code but LLMs can hallucinate or introduce subtle bugs; (3) be mindful of copyright/licensing when the skill copies examples from external sources — require attribution or rewrite as needed; (4) if your environment disallows outbound web access, confirm the agent's data sources (the SKILL.md assumes ability to search docs/blogs); and (5) ask the skill to list sources and, when needed, shorten or alter reproduced material to avoid verbatim copyrighted text. Overall the skill appears coherent and proportionate, but review outputs for technical accuracy and licensing before publishing.Like a lobster shell, security has layers — review code before you run it.
latestvk9703j2k6fey4nf2kd2ytr9t2584efbh
License
MIT-0
Free to use, modify, and redistribute. No attribution required.