solidity-security
Pass
Audited by Gen Agent Trust Hub on Oct 23, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill references security research and audit reports from OpenAI and Code4rena to inform its analysis criteria.
- [COMMAND_EXECUTION]: The skill suggests running standard development tools such as forge, cast, and slither to verify code security and performance.
- [INDIRECT_PROMPT_INJECTION]: The skill analyzes user-provided Solidity source code, which constitutes an ingestion surface for potential instructions.
- Ingestion points: Solidity (.sol) files.
- Capability inventory: Execution of developer tools like forge and slither.
- Boundary markers: Not explicitly defined.
- Sanitization: Relies on agent analysis of code patterns.
- [SAFE]: The instructions explicitly forbid the agent from reading or exposing sensitive environment files and private keys, providing safer alternatives like local keystores.
Audit Metadata