web3-expert

Pass

Audited by Gen Agent Trust Hub on Apr 25, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: Indirect Prompt Injection Surface. The skill is designed to review and refactor code, which acts as an ingestion point for untrusted external data. Because the skill has access to sensitive tools (Bash, Write, Edit), it is theoretically vulnerable if the reviewed code contains malicious instructions that the agent might inadvertently follow.
  • Ingestion points: The skill accepts code for review from user queries as described in SKILL.md.
  • Boundary markers: There are no explicit instructions or delimiters used to ensure the agent distinguishes between the code being reviewed and the instructions it should follow.
  • Capability inventory: The skill utilizes powerful tools including Bash, Write, Edit, Grep, and Glob as listed in the SKILL.md frontmatter.
  • Sanitization: No input validation or sanitization mechanisms are implemented to filter out embedded instructions in the code being analyzed.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 25, 2026, 04:16 PM
Security Audit — agent-trust-hub — web3-expert