account-abstraction

Pass

Audited by Gen Agent Trust Hub on Sep 5, 2026

Risk Level: SAFE
Full Analysis
  • [COMMAND_EXECUTION]: The skill mentions the use of cairo-auditor as a step in its workflow. This is a domain-specific auditing tool for Cairo/Starknet and is used here in a professional security context. The skill itself does not automatically execute this tool but instructs the user to do so.
  • [DATA_EXPOSURE]: The starknet.js code example demonstrates the usage of process.env.PRIVATE_KEY. This is documented as a standard and safe practice for secret management, specifically advising the use of environment variables rather than hardcoded credentials.
  • [REMOTE_CODE_EXECUTION]: The skill refers to cairo-auditor and starknet.js (an NPM package). These are standard tools within the Starknet development ecosystem and are handled as safe, context-appropriate references.
  • [INDIRECT_PROMPT_INJECTION]: The skill includes a surface for reviewing account abstraction flows, which involves processing external code or configurations. However, it lacks dangerous autonomous capabilities (like writing to sensitive files or making arbitrary network requests based on that data) and focuses on providing human-in-the-loop guidance.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 5, 2026, 06:31 AM
Security Audit — agent-trust-hub — account-abstraction