persona-researcher

Pass

Audited by Gen Agent Trust Hub on Jul 15, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: The skill defines a persona that processes external research papers and notes, providing a surface for indirect prompt injection.
  • Ingestion points: Research documents and notes managed through gws-drive and gws-docs (SKILL.md).
  • Boundary markers: The instructions lack specific delimiters or warnings to prevent the agent from executing instructions potentially embedded within the research materials.
  • Capability inventory: The persona can perform sensitive actions such as sending emails (gws gmail +send), writing to documents (gws docs +write), and appending to spreadsheets (gws sheets +append) (SKILL.md).
  • Sanitization: There are no mentions of sanitizing or validating external document content to mitigate potential injection attacks.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 15, 2026, 01:27 PM
Security Audit — agent-trust-hub — persona-researcher