eigenlayer

Pass

Audited by Gen Agent Trust Hub on Aug 4, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill primarily consists of educational content and code templates for interacting with the EigenLayer protocol on Ethereum. It includes contract addresses that match official documentation for the protocol.
  • [DATA_EXPOSURE]: All code examples for transaction signing (e.g., in templates/restaking-client.ts and SKILL.md) correctly implement secret management by referencing environment variables (process.env.PRIVATE_KEY) rather than hardcoding credentials.
  • [EXTERNAL_DOWNLOADS]: The skill references standard, well-known libraries such as viem and @openzeppelin/contracts. No suspicious or unverified third-party dependencies were found.
  • [PROMPT_INJECTION]: The instructional content is purely technical and does not contain patterns intended to bypass agent safety filters or override system instructions.
  • [COMMAND_EXECUTION]: Shell commands provided in resources/contract-addresses.md (using cast) are intended for manual user verification of on-chain contract code and do not involve autonomous execution of dangerous commands.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 4, 2026, 12:30 AM
Security Audit — agent-trust-hub — eigenlayer