lp-engineer
Pass
Audited by Gen Agent Trust Hub on Sep 4, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to interact with external, untrusted data sources such as blockchain RPC nodes, indexers, and protocol-specific APIs.
- Ingestion points: Data from repositories, RPC, APIs, and token hooks (SKILL.md).
- Boundary markers: Instructions explicitly state that such data must be treated as untrusted data rather than agent instructions.
- Capability inventory: The skill is capable of constructing transactions and performing read-only fork tests (SKILL.md, references/test-matrix.md).
- Sanitization: Emphasizes typed snapshots and explicit result types (references/adapter-architecture.md) to prevent malformed data from affecting control flow.
Audit Metadata