daily-report
v1.0.2获取每日日报，通过 API 获取用户新闻推荐信息，支持用户兴趣偏好和查询关键词排序，获取个性化每日日报，并生成精美 H5 页面。
⭐ 0· 44·0 current·0 all-time
MIT-0
Download zip
LicenseMIT-0 · Free to use, modify, and redistribute. No attribution required.
Security Scan
OpenClaw
Benign
medium confidencePurpose & Capability
Name/description, included Python scripts, example data, and output HTML are coherent: the skill fetches news from the declared API, allows model-driven personalization, and generates an H5 report. No unrelated credentials, binaries, or configs are requested.
Instruction Scope
SKILL.md explicitly instructs the LLM (the agent) to read the local data/daily_report.json, use the current conversation context and platform memory (灵犀记忆) for personalization, produce a sorted ID list and save it to data/sorted_ids.json (deleting any existing file first). These actions are within the stated personalization purpose but require the agent to expose conversation/memory-derived preferences into a local JSON file (sorting_reason) — a privacy consideration.
Install Mechanism
No install spec; the package includes Python scripts that will be executed. No external installers, downloads, or extracted archives are present in the manifest.
Credentials
The skill requests no environment variables or credentials, which is consistent. It does make an outbound HTTP(S) request to https://jiutian.10086.cn/jiujiuassist/proactive/get_user_news_recommend (declared in SKILL.md and used by get_daily_report.py) — network access is required but no secrets are asked for.
Persistence & Privilege
always is false and the skill does not request system-wide configuration changes or persistent elevated privileges. It writes files under the skill package directories (data/ and output/) only.
Assessment
What to check before installing or running: 1) Network & trust: the script calls https://jiutian.10086.cn — verify you trust that endpoint and that outbound HTTP access is acceptable. 2) Privacy: the LLM sorting step is required to use current conversation context and platform memory (灵犀记忆) and will write a human-readable sorting_reason to data/sorted_ids.json; that may persist user-preference information locally. 3) Files: the skill will create/overwrite files under the skill's data/ and output/ directories and asks the agent to delete an existing sorted_ids.json — ensure you’re comfortable with that file operation. 4) Dependencies: install requests in the execution environment (pip install requests). 5) If you have sensitive conversation memory or operate in a locked-down/network-restricted environment, review or sandbox execution and inspect the included scripts (they are provided) before running.Like a lobster shell, security has layers — review code before you run it.
latestvk976nv4f3tn7svdwdneyc551n5844pgm
License
MIT-0
Free to use, modify, and redistribute. No attribution required.