review-paper

Pass

Audited by Gen Agent Trust Hub on Jun 22, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: Indirect Prompt Injection Surface\n
  • Ingestion points: The skill accepts user-provided research topics, format preferences, and descriptions of prior/adjacent reviews (SKILL.md Step 0, skill.yml inputs).\n
  • Boundary markers: Absent; user input and referenced literature content are interpolated into manuscript scaffolds without delimiters or warnings to ignore embedded instructions.\n
  • Capability inventory: The skill uses Bash, Write, Edit, and Glob tools to create and modify manuscript files on the local system.\n
  • Sanitization: There is no documented validation or sanitization of the input provided by the user or extracted from research sources before it is processed by the agent.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 22, 2026, 03:56 AM
Security Audit — agent-trust-hub — review-paper