public-relations

Pass

Audited by Gen Agent Trust Hub on Jun 29, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill fetches news data from well-known services (Google News, Reddit, and Algolia) using curl to identify trending topics. These references are used for legitimate newsjacking purposes.
  • [COMMAND_EXECUTION]: The skill provides shell command examples for using curl, xmllint, and jq to process news feeds. These are standard developer tools used appropriately.
  • [PROMPT_INJECTION]: The skill processes untrusted headlines from external news sources. While this introduces a surface for indirect prompt injection, it is assessed as a low-risk inherent property of the news-monitoring functionality.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 29, 2026, 01:08 PM