learning-literature-review

Pass

Audited by Gen Agent Trust Hub on Feb 17, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [Indirect Prompt Injection] (LOW): The skill ingests and processes literature data which could contain malicious hidden instructions.
  • Ingestion points: The input_data dictionary in skill.py is the primary entry point for untrusted external content.
  • Boundary markers: There are no delimiters or explicit instructions provided to the underlying model to ignore instructions embedded within the research data.
  • Capability inventory: The skill is capable of generating reports and identifying insights; while system-level access is limited in the provided snippet, the logic relies on a local framework.
  • Sanitization: No input validation or escaping mechanisms are applied to the input_data before use.\n- [Code Quality] (SAFE): The variable skill_dir used in the execute method in skill.py is not defined within the function or imported, which will result in a NameError during execution.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 17, 2026, 06:21 PM