press-release-writing

Pass

Audited by Gen Agent Trust Hub on Mar 11, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill instructs the user to use the infsh CLI for authentication and running research applications, which is standard functionality for the provided toolset.
  • [EXTERNAL_DOWNLOADS]: Recommends the installation of additional tools from the inference-sh organization using npx. This involves fetching external code from a vendor-controlled source.
  • [PROMPT_INJECTION]: Vulnerability surface for indirect prompt injection exists due to the processing of external search data.
  • Ingestion points: Results from tavily/search-assistant and exa/search are used to populate the press release content.
  • Boundary markers: No delimiters or specific instructions are provided to the agent to ignore instructions embedded in the search results.
  • Capability inventory: The skill utilizes Bash(infsh *) for tool execution.
  • Sanitization: No data sanitization or validation steps are defined for the external input.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 11, 2026, 05:20 PM