expense-categorization
Pass
Audited by Gen Agent Trust Hub on Sep 17, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes external, untrusted transaction data which could contain malicious instructions designed to influence the agent's behavior.
- Ingestion points: The skill accepts transaction data in CSV or text formats as described in the 'Receive Expense Data' workflow step in SKILL.md.
- Boundary markers: None identified; the instructions do not include specific delimiters or directions to the agent to ignore instructions embedded within the processed data.
- Capability inventory: No dangerous capabilities such as network access, shell execution, or file writing were identified in the skill instructions or metadata.
- Sanitization: The skill does not define any sanitization, filtering, or validation logic to inspect the content of the transaction descriptions for malicious payloads.
Audit Metadata