rental-property
Pass
Audited by Gen Agent Trust Hub on Sep 15, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted external data in the form of transaction descriptions and notes retrieved via transaction_search.
- Ingestion points: External transaction descriptions and vendor names are ingested into the agent context via transaction_search in the Workflow (Step 2 and 3) in SKILL.md.
- Boundary markers: The instructions do not define delimiters or specific warnings to ignore instructions embedded within the transaction data.
- Capability inventory: The skill uses financial analysis tools (transaction_search, spending_summary, export_transactions) but lacks high-risk capabilities like shell execution or arbitrary file writes.
- Sanitization: No explicit sanitization or filtering of transaction descriptions is mentioned.
Audit Metadata