defi-attack-patterns
Pass
Audited by Gen Agent Trust Hub on Jul 21, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill defines a decision tree for analyzing external DeFi protocols, which involves ingesting untrusted data such as smart contract code and protocol parameters. This creates a potential surface for indirect prompt injection if the analyzed content contains malicious instructions.
- Ingestion points: Section 9 ('Decision Tree') directs the agent to evaluate external protocol components during analysis.
- Boundary markers: Absent. There are no instructions to wrap or delimit analyzed protocol data to prevent the agent from executing instructions embedded within it.
- Capability inventory: The skill focus is instructional analysis; it does not explicitly invoke shell commands, file writes, or network operations within the SKILL.md content.
- Sanitization: Absent. The instructions do not include steps to sanitize or validate the external protocol data before it is processed by the agent.
Audit Metadata