money-review-skeptic

Pass

Audited by Gen Agent Trust Hub on May 5, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill is an instructional prompt for business plan red-teaming. It does not exhibit malicious behavior, use obfuscated code, or attempt unauthorized system modifications.
  • [PROMPT_INJECTION]: The skill ingests external data from business plans and web searches, creating an indirect prompt injection surface. 1. Ingestion points: Business plans and web search results. 2. Boundary markers: Absent. 3. Capability inventory: Web search and file reading. 4. Sanitization: None. This is a common characteristic of analysis-oriented skills.
  • [DATA_EXFILTRATION]: The skill accesses local conversation context and user-specified files to perform its review. This behavior is consistent with its stated purpose and does not involve sending data to untrusted remote destinations.
Audit Metadata
Risk Level
SAFE
Analyzed
May 5, 2026, 05:47 PM