yield-optimizer
Pass
Audited by Gen Agent Trust Hub on Mar 11, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: No malicious code, hardcoded credentials, or suspicious network behavior was found in the skill definitions. The skill implements robust security practices, including automated protocol risk scoring and token security audits.
- [PROMPT_INJECTION]: The skill has an indirect prompt injection surface because it ingests untrusted data from external DeFi protocols. An attacker could attempt to influence the agent by placing instructions in protocol names or metadata on platforms like DeFiLlama.
- Ingestion points: External protocol and market data fetched via DeFiLlama MCP, CoinGecko MCP, and GoPlus MCP (SKILL.md).
- Boundary markers: Absent; the instructions do not explicitly tell the model to ignore instructions embedded in the external metadata.
- Capability inventory: Limited to read-only data fetching and analysis; trade execution and wallet operations are explicitly excluded (SKILL.md).
- Sanitization: The skill mitigates risks by requiring automated token security checks via the GoPlus MCP before recommending any DeFi opportunity.
Audit Metadata