skills/booch/config_files/ground-news/Gen Agent Trust Hub

ground-news

Pass

Audited by Gen Agent Trust Hub on Jun 16, 2026

Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
  • [PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface by ingesting untrusted data from multiple external sources.\n
  • Ingestion points: Fetches content from Ground News API, RSS feed URLs, and IMAP email digests (SKILL.md).\n
  • Boundary markers: No explicit delimiters or instructions to ignore embedded commands are present in the skill definition.\n
  • Capability inventory: The processed content is aggregated into a morning briefing and converted to HTML format for user consumption.\n
  • Sanitization: No sanitization, escaping, or validation of the external content is described.\n- [COMMAND_EXECUTION]: The skill invokes the 1Password CLI to manage credentials.\n
  • Evidence: Executing op read to access the Ground News API key stored in a 1Password vault (SKILL.md). While this is a secure method for handling secrets, it involves executing shell commands to interact with local software.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 16, 2026, 10:20 PM
Security Audit — agent-trust-hub — ground-news