academy-of-management-journal

Pass

Audited by Gen Agent Trust Hub on Jun 24, 2026

Risk Level: SAFENO_CODE
Full Analysis
  • [DATA_EXPOSURE]: The skill instructs the agent to read configuration or reference data from relative paths outside its own directory, specifically ../../resources/source-basis.md and ../../resources/official-source-map.md. While these are project-internal resources, they represent an access pattern to shared data files.
  • [INDIRECT_PROMPT_INJECTION]: The skill acts as a processing layer for user-provided research manuscripts. This creates a surface for indirect prompt injection if a manuscript contains instructions designed to manipulate the agent's behavior.
  • Ingestion points: User-provided manuscripts and research abstracts.
  • Boundary markers: None identified.
  • Capability inventory: No subprocesses, network writes, or file-write capabilities identified; the skill only generates text output.
  • Sanitization: No input validation or escaping mechanisms are defined in the instructions.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 24, 2026, 04:01 AM
Security Audit — agent-trust-hub — academy-of-management-journal