ClawHub

AI日报测试版

v1.0.0

专门用于整理、汇总和深度分析每日最前沿的 AI 资讯，包括行业大事件、大模型更新、学术论文、开源工具及新技术标准。当用户提到“生成 AI 日报”、“今日 AI 资讯”、“行业动态汇总”或类似的请求时，请务必使用此技能。即使没有明确说“日报”，只要是请求汇总最新 AI 领域进展，都应触发本技能以确保内容的深度和准确性。

⭐ 0· 70·0 current·0 all-time
by@ranbin
MIT-0
Download zip
LicenseMIT-0 · Free to use, modify, and redistribute. No attribution required.
Security Scan
VirusTotalVirusTotal
Benign
View report →
OpenClawOpenClaw
Benign
high confidence
✓
Purpose & Capability
The name/description match the SKILL.md instructions: both specify bilingual web retrieval, source lists (OpenAI, arXiv, GitHub, tech press, etc.), classification, and deep analysis. There are no unexpected env vars, binaries, or config paths requested.
✓
Instruction Scope
Runtime instructions are narrowly scoped to web searches, content classification, analysis, and Markdown output. The SKILL.md explicitly directs use of a web search tool (search_web) and lists expected sources; it does not instruct reading local files, harvesting environment variables, or posting collected data to unrelated external endpoints. Monitor for potential hallucinated source links (common risk when summarizing web results).
✓
Install Mechanism
No install spec and no code files — instruction-only. Nothing will be written to disk or downloaded as part of skill installation.
✓
Credentials
No environment variables, credentials, or config paths are requested. The absence of secrets is proportionate to the described functionality.
✓
Persistence & Privilege
always is false and the skill is user-invocable. Autonomous invocation is allowed by platform default but the skill does not request elevated or persistent privileges.
Assessment
This skill appears coherent and safe to install: it only instructs the agent to search and summarize public AI news sources and requests no credentials. Before relying on outputs, verify a few things: (1) confirm the platform's search_web tool is available and configured as you expect; (2) spot-check the cited links and timestamps to catch hallucinated or stale references; (3) if you need trusted sources only, consider asking for a source whitelist in the prompt; and (4) monitor the skill's first runs to ensure it doesn't request extra permissions or try to access non-public data.

Like a lobster shell, security has layers — review code before you run it.

latestvk9726h1pymp3agetph960pvwnd842vyn

License

MIT-0
Free to use, modify, and redistribute. No attribution required.
Termshttps://spdx.org/licenses/MIT-0.html

Comments