Skill flagged — suspicious patterns detected
ClawHub Security flagged this skill as suspicious. Review the scan results before using.
热点聚合监控
v1.0.0🔥 热点聚合监控 - 一站式聚合微博/百度/知乎/抖音热搜榜，自动生成每日热点报告，支持关键词订阅推送。适用于自媒体运营、内容创作、市场分析等场景。
⭐ 0· 122·0 current·0 all-time
MIT-0
Download zip
LicenseMIT-0 · Free to use, modify, and redistribute. No attribution required.
Security Scan
OpenClaw
Suspicious
medium confidencePurpose & Capability
Name/description match the code: scripts fetch platform hot-search data, generate a markdown report, and manage keyword subscriptions. However, package.json lists required bins (curl, jq) while registry metadata claimed none; scripts also write/read under /root/clawd/memory which is not declared. These mismatches are disproportionate to the stated purpose (should not assume root paths or omit required binaries).
Instruction Scope
SKILL.md and the scripts instruct the agent to call external URLs (weibo, baidu, zhihu, and a third-party aggregator api.oioweb.cn), create and read files under /root/clawd/memory/hotspots, and use environment variables USE_REAL_API and PROXY although the skill declared no required env vars. The scripts will fetch remote data and write persistent files outside the skill directory — these behaviors are within the skill's stated purpose but the undocumented path and env usage are scope-creep that should be clarified.
Install Mechanism
No install spec (instruction-only with shipped scripts). No downloads from external URLs during install; code is bundled with the skill. This is low install-risk.
Credentials
The skill declares no required env vars or credentials but runtime uses optional envs USE_REAL_API and PROXY and package.json implies dependency on curl and jq. It also calls external third-party API endpoints (api.oioweb.cn). No secrets are requested, but the undeclared env/binary expectations and external endpoints are disproportionate to the manifest and should be documented.
Persistence & Privilege
Skill writes persistent data to /root/clawd/memory/hotspots and creates/edits config.json next to the scripts. Hard-coding a /root path is a privilege/assumption mismatch (may require root or a specific runtime layout). always:false (no forced global presence) so autonomous invocation is normal, but the skill's use of system/root paths increases its blast radius and should be addressed.
What to consider before installing
This skill appears to implement its advertised features, but there are clear inconsistencies you should resolve before installing: (1) package.json expects curl and jq, and the runtime uses optional env vars USE_REAL_API and PROXY — these are not declared in the registry metadata; ensure your environment provides curl/jq and understand the effect of setting USE_REAL_API=true. (2) Scripts write/read data under /root/clawd/memory/hotspots (and create/edit config.json relative to the skill) — change DATA_DIR to a non-root path or run in a contained environment to avoid unintended writes. (3) The skill calls external endpoints (weibo/zhihu/baidu and third-party api.oioweb.cn). If you enable real API mode, network requests will go to those servers; review the endpoints and your network/ privacy policy. (4) Rate limits and proxy needs are mentioned — configure a proxy or respect API limits. Recommended actions: run the skill in a sandbox or non-privileged account, inspect and if needed modify DATA_DIR and config paths, verify/whitelist external endpoints, and ensure required binaries (curl, jq) are installed. If you need higher assurance, ask the author to update the manifest to declare required binaries, env vars, and the data path.Like a lobster shell, security has layers — review code before you run it.
latestvk97d0tbkf4dzzsgjqbnmhdjeyn83b2pp
License
MIT-0
Free to use, modify, and redistribute. No attribution required.
Runtime requirements
🔥 Clawdis