mythril
Pass
Audited by Gen Agent Trust Hub on Aug 4, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill primarily serves as a documentation and configuration resource for Mythril, a legitimate and widely-used security analysis tool for smart contracts.
- [EXTERNAL_DOWNLOADS]: The skill fetches configuration and software from official, trusted sources including Docker Hub (mythril/myth), PyPI, and the Consensys GitHub organization. These references are documented neutrally and align with standard development practices.
- [COMMAND_EXECUTION]: The provided shell scripts and GitHub Action workflows are designed to automate security scans. They include defensive configurations such as CPU/Memory limits and multi-layer timeouts to prevent resource exhaustion or hanging processes in automated environments.
- [DATA_EXFILTRATION]: Analysis of the network-related examples shows no evidence of malicious data routing. RPC URL examples correctly use environment variable placeholders or standard public endpoints without attempting to harvest local environment secrets.
- [PROMPT_INJECTION]: The instructional content is technical and focuses on tool usage. It contains no patterns of adversarial role-play, DAN-style overrides, or instructions intended to bypass the AI agent's safety guardrails.
Audit Metadata