Skill flagged — suspicious patterns detected
ClawHub Security flagged this skill as suspicious. Review the scan results before using.
Morning Briefing Generator 晨会简报生成器
v1.0.0每日早会简报生成器 — 自动聚合国内外新闻、财经资讯、行业动态，生成结构化早会简报并推送给指定用户。适用于企业高管、创业者、投资者等需要快速掌握每日要闻的场景。支持定时任务自动执行。
⭐ 0· 43·0 current·1 all-time
MIT-0
Download zip
LicenseMIT-0 · Free to use, modify, and redistribute. No attribution required.
Security Scan
OpenClaw
Suspicious
medium confidencePurpose & Capability
The README and SECURITY.md reference a scripts/generate-briefing.ps1 implementation and a code-based workflow, but the published bundle contains only documentation files (SKILL.md, README*, SECURITY.md) and no scripts. The skill advertises 'push to specified users' and cron scheduling but declares no credentials, endpoints, or mechanism for pushing notifications. This inconsistency between claimed capabilities (scripted generation and push) and the actual package contents is unexpected.
Instruction Scope
SKILL.md's metadata permits network access to wttr.in and api.allorigins.win and allows reading/writing the workspace; that fits news/weather fetching. However SKILL.md also repeatedly claims 'no external data transmission' and 'no credentials', which is ambiguous: fetching web pages is external I/O (acceptable for news), but the unspecified 'push' step could require transmitting generated content or credentials. The instructions are vague about where/how push happens and what data (workspace contents) might be sent.
Install Mechanism
No install spec and no code files are present, so there is no installer or remote download to evaluate — this is low install risk. However the documentation's references to a script that isn't included are misleading.
Credentials
The skill declares no required environment variables or credentials, which is proportionate to an offline/local generator that only writes to workspace. But because the skill promises scheduled generation and 'push' to users, the absence of any declared credential or config for notification services (email/Slack/WeChat) is a mismatch that should be resolved before trusting automated pushes.
Persistence & Privilege
The skill is not marked always:true and uses default invocation rules. It requests only optional workspace read/write and limited network access in metadata. There is no indication the package will force persistent system-level changes.
What to consider before installing
Do not enable automated scheduling or allow the skill to push content until the missing pieces are clarified. Specific checks to request from the publisher: (1) supply the actual generate script(s) referenced in README/SECURITY (scripts/generate-briefing.ps1) so you can review them; (2) explain exactly how "push to specified users" works and list endpoints, required credentials, or integrations (email, Slack, webhook, etc.); (3) confirm the exact network endpoints the skill will call at runtime and whether any workspace content could be transmitted to those endpoints; (4) verify the origin/identity of the publisher and ask why the Security audit and script are bundled only as claims in docs but not present in the package. If you must test it, run it in a restricted sandbox with no real credentials and with workspace contents you consider non-sensitive. If the author cannot provide the missing script/source or a clear explanation of push mechanics, treat the package as untrusted.Like a lobster shell, security has layers — review code before you run it.
latestvk97fgcqs1kvbcc48rz78c7v14d84qn0x
License
MIT-0
Free to use, modify, and redistribute. No attribution required.
Runtime requirements
🌅 Clawdis