real-estate-analyst

Pass

Audited by Gen Agent Trust Hub on Aug 14, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill uses a multi-agent orchestration pattern to perform real estate analysis tasks. It manages data flow between specialized agents (market, location, profitability, risk) using a local workspace for intermediate files.
  • [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted user data, including property addresses and listings, which is a surface for indirect prompt injection. However, the agents' actions are limited to generating analytical reports and do not involve executing code or accessing sensitive system resources.
  • Ingestion points: User-provided property details and investment parameters are extracted and saved to _workspace/00_input.md in SKILL.md.
  • Boundary markers: Absent; user input is organized into markdown without specific delimiters or instructions to ignore embedded commands.
  • Capability inventory: File system writes (Markdown files), inter-agent communication via SendMessage, and task tracking via TaskCreate and TaskUpdate.
  • Sanitization: No sanitization or validation of the content of user-provided property descriptions is specified.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 14, 2026, 01:10 PM
Security Audit — agent-trust-hub — real-estate-analyst