due-diligence

Pass

Audited by Gen Agent Trust Hub on Nov 11, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTIONNO_CODE
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill is designed to process and analyze untrusted external data sources, which presents a surface for indirect prompt injection attacks.\n
  • Ingestion points: Virtual data rooms, customer records, and legal contracts as described in SKILL.md and references/commercial-operational.md.\n
  • Boundary markers: The instructions do not specify the use of delimiters or instructions to ignore embedded prompts within the external documents.\n
  • Capability inventory: The skill instructs the agent to perform research, reach conclusions, and potentially use restricted systems, which are actions that could be manipulated by adversarial content in the ingested files.\n
  • Sanitization: No methods for sanitizing or filtering external content are mentioned in the instructions.\n- [NO_CODE]: The skill consists exclusively of documentation and instructional Markdown files with no associated scripts or binaries, eliminating risks associated with direct code execution.
Audit Metadata
Risk Level
SAFE
Analyzed
Nov 11, 2026, 10:05 PM