deal-screening

Pass

Audited by Gen Agent Trust Hub on May 22, 2026

Risk Level: SAFE
Full Analysis
  • Indirect Prompt Injection Surface: The skill is intended to ingest and analyze external deal materials such as CIMs and teasers. Because these documents are provided by external parties, they represent a potential surface for indirect prompt injection where text could be crafted to influence the agent's evaluation.
  • Ingestion points: SKILL.md directs the agent to process provided CIMs, teasers, or descriptions provided by the user.
  • Boundary markers: No specific delimiters or "ignore embedded instructions" warnings are used to separate the deal content from the agent's instructions, which is a common pattern in document analysis skills.
  • Capability inventory: The skill's operations are strictly limited to text extraction, criteria comparison, and memo generation; it does not utilize network requests, file system modifications, or shell execution capabilities.
  • Sanitization: The skill does not explicitly sanitize the content of the provided deal materials before processing, relying on the agent's internal safety guardrails.
Audit Metadata
Risk Level
SAFE
Analyzed
May 22, 2026, 11:02 AM
Security Audit — agent-trust-hub — deal-screening