wizard

Pass

Audited by Gen Agent Trust Hub on Aug 13, 2026

Risk Level: SAFECOMMAND_EXECUTIONCREDENTIALS_UNSAFE
Full Analysis
  • [DYNAMIC_EXECUTION]: The skill generates shell scripts for manual procedures and instructs the user to execute them locally via chmod +x.
  • [INDIRECT_PROMPT_INJECTION]: The agent determines setup steps by reading untrusted files from the repository.
  • Ingestion points: .env, README, docker-compose, and workflow files as specified in SKILL.md.
  • Boundary markers: None specified to separate repo content from script generation instructions.
  • Capability inventory: The generated scripts can open URLs, write to local environment files, and modify GitHub secrets using template.sh helpers.
  • Sanitization: No explicit sanitization or validation of the ingested repository data is implemented.
  • [DATA_EXPOSURE]: The skill is specifically designed to capture and persist sensitive credentials in local environment files and GitHub secrets.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 13, 2026, 03:09 AM
Security Audit — agent-trust-hub — wizard