reports

Pass

Audited by Gen Agent Trust Hub on Aug 29, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill provides a mechanism for the agent to process and analyze external content like briefings and research reports to validate embedded links. This creates a surface for indirect prompt injection where the agent ingests untrusted data from the web. Ingestion points: External content via tools like search and get_page. Boundary markers: None explicitly mentioned in the instructions. Capability inventory: File writing, network access, and repository status checks. Sanitization: No specific filtering or sanitization logic is described.
  • [COMMAND_EXECUTION]: The skill instructions direct the agent to execute the 'git status' command within the brain repository to verify that linked files are committed and pushed.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 29, 2026, 06:18 PM
Security Audit — agent-trust-hub — reports