University Applications
v1.0.0采集香港所有大学（含公立及私立共22所院校）的官方硕士项目招生信息， 包括学费、申请时间、截止日期、英语要求、项目详情及官方链接， 并输出为 Excel、Word、PDF、HTML 和 Markdown 五种格式。 仅使用大学官方数据源，严禁任何第三方信息。
⭐ 0· 35·0 current·0 all-time
byenoyao@wscats
MIT-0
Download zip
LicenseMIT-0 · Free to use, modify, and redistribute. No attribution required.
Security Scan
Capability signals
These labels describe what authority the skill may exercise. They are separate from suspicious or malicious moderation verdicts.
OpenClaw
Benign
high confidencePurpose & Capability
The skill's name/description (collect HK master's admissions info and export to Excel/Word/PDF/HTML/Markdown) matches the included files: SKILL.md, data schema, per-university prompts, output templates, and sample outputs. No extraneous binaries, credentials, or unrelated capabilities are requested.
Instruction Scope
Runtime instructions are narrowly scoped to: read the spec/schema, visit official university pages listed in the universities/ files, extract the specified 20 fields per program, validate formats, and apply templates. This is coherent. Operational note: the prompts assume the agent can fetch/visit official web pages; if the executing agent lacks browsing/web access the LLM may hallucinate data—this is a functional (not security) risk to be aware of.
Install Mechanism
No install spec or code is included; the skill is pure Markdown. Nothing will be written to disk or executed by an installer as part of install. This minimizes installation risk.
Credentials
The skill requests no environment variables, credentials, or config paths. All required data sources are explicit official domains. There are no unexplained secret or credential requests.
Persistence & Privilege
Skill flags indicate default invocation behavior (always:false, model invocation enabled) which is normal. The skill does not request permanent presence or system-wide config changes and does not attempt to modify other skills.
Assessment
This skill appears coherent and low-risk: it contains only prompts, templates, and example outputs and requests no credentials. Before installing, confirm that (1) your agent has appropriate web/browsing access if you expect live scraping; without it the model could fabricate details, (2) you will manually verify critical data against the provided official URLs (the skill itself instructs verification), and (3) you are comfortable allowing the agent to make web requests when invoked. If you prefer tighter control, restrict autonomous invocation or manually run the workflows so you can validate results before sharing or storing them.Like a lobster shell, security has layers — review code before you run it.
latestvk977dwaq7pgxc53av6q6jrkjj5849gf5
License
MIT-0
Free to use, modify, and redistribute. No attribution required.