eth-to-sol
Pass
Audited by Gen Agent Trust Hub on Aug 9, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill ingests untrusted Solidity source code from users to generate Solana program translations, creating an inherent surface for indirect prompt injection where malicious instructions in a contract could attempt to influence the agent's behavior. Ingestion points: User-supplied Solidity files or code snippets. Boundary markers: The skill does not specify delimiters (such as XML tags) to isolate untrusted input from system instructions. Capability inventory: The skill can generate and modify workspace files and call the rust_autofixer tool from the Solana Developer MCP. Sanitization: No sanitization or filtering is applied to the input Solidity code before processing.
Audit Metadata