press-release-writing

Pass

Audited by Gen Agent Trust Hub on Aug 31, 2026

Risk Level: SAFE
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill references the inference.sh (belt) CLI tool and provides instructions to install it from a GitHub source. This is documented neutrally as it relates to the skill's primary function of fact-checking and research using established tools.
  • [COMMAND_EXECUTION]: The skill includes documentation for using the belt CLI tool to run search queries via Tavily and Exa. These are standard search assistant operations used for fact-checking press release claims and providing industry context.
  • [INDIRECT_PROMPT_INJECTION]: The skill processes user-provided inputs for press release topics (e.g., funding rounds, product launches). It mitigates risks by providing structured templates and guidelines that enforce professional AP style formatting, reducing the surface for unintended instruction execution.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 31, 2026, 11:00 PM
Security Audit — agent-trust-hub — press-release-writing