Security audit
Security checks for vulnerabilities and agentic risk
This skill is a straightforward AI news digest helper that searches current web sources and can optionally save a markdown newsletter when requested.
Install this if you want an agent to browse current AI news and produce concise digests. Use inline output if you do not want files created, and choose file output only when you are comfortable with a dated newsletter being written under the outputs directory.
The invocation guidance says to use this skill whenever a user asks for AI news, wants a roundup, asks what happened today or this week, or even says follow-ups like 'more AI news' or 'catch me up on AI.' These broad phrases overlap with ordinary conversational requests and explicitly expand activation even when the user does not ask for a newsletter, increasing the risk of unintended invocation.
Skill enables autonomous high-impact decisions without human-in-the-loop verification. Critical operations (destructive commands, financial transactions, data deletion) should require explicit user confirmation.
that differ from the defaults. One question, grouped options — don't
interrogate them.
**Defaults (proceed without asking if user gave enough context):**
- Article count: 10
- Time window: last 2 days
- Topic focus: all AI/ML (no filter)
The skill instructs the agent to write a file into user storage and present it, but it does not require an explicit confirmation or warning before creating the file. While the path appears constrained and the filename is date-based, silent file creation can still surprise users, create clutter, or overwrite prior outputs if rerun on the same date.
No suspicious patterns detected.