News Brief Automation
v1.0.0Build, refine, or maintain recurring news-digest workflows that periodically collect items from sources such as Hacker News, GitHub Trending, Hugging Face, R...
⭐ 0· 50·0 current·0 all-time
bylouis@louishwh
MIT-0
Download zip
LicenseMIT-0 · Free to use, modify, and redistribute. No attribution required.
Security Scan
OpenClaw
Benign
high confidencePurpose & Capability
Name/description (news-digest automation) match the SKILL.md content: creating cron prompt files, report folders, dedupe rules, saving Markdown, and sending summaries. There are no unrelated requirements (no env vars, no binaries).
Instruction Scope
Runtime instructions are scoped to identifying sources, selecting items, deduplicating against the newest report in the specified reports/ folder, saving a Markdown file, and sending that file to the user. All referenced files and paths (automation/cron/, reports/) align with the described purpose. The instructions do imply web fetching of public sources (HN, GitHub, Hugging Face), which is expected for this task and not performed by the skill itself (it's guidance for the agent).
Install Mechanism
Instruction-only skill with no install spec and no code files. Nothing is downloaded or written by an installer; lowest-risk installation posture.
Credentials
No required environment variables, credentials, or config paths are declared. The tasks described (reading/writing reports in workspace, optionally fetching public webpages) don't justify any additional secrets or system access.
Persistence & Privilege
always:false and normal model invocation are used. The skill does not request persistent system-wide privileges or attempt to modify other skills' configs. Its save/read behavior is limited to the workspace paths it instructs the agent to use.
Assessment
This skill is a set of instructions/templates for building recurring news-digest automations and appears internally consistent. Before installing or enabling it, confirm: (1) you are comfortable with the agent reading and writing files under automation/cron/ and reports/ in the workspace; (2) any delivery target (email, chat, webhook) you configure for 'send after save' is intentionally provided by you — do not embed credentials in prompt files; (3) automated runs that fetch public sites will access external URLs (HN, GitHub, Hugging Face) — ensure your environment's network policy allows only intended outbound access; and (4) review the prompt templates/examples to avoid accidentally publishing sensitive or private links. No additional secrets or installs are required by this skill.Like a lobster shell, security has layers — review code before you run it.
latestvk9721ebvs9bwfpm8fwb34164n183wjvt
License
MIT-0
Free to use, modify, and redistribute. No attribution required.