aidlc-reverse-engineer

Pass

Audited by Gen Agent Trust Hub on Jul 20, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill is designed for brownfield codebase analysis, requiring access to local source files to generate documentation. All operations are confined to reading the file system and writing reports to the .aidlc/reverse-engineer/ directory.
  • [COMMAND_EXECUTION]: The skill uses standard utilities like wc -l to calculate lines of code. This is a common and expected operation for a tool analyzing project statistics and does not pose a security risk in this context.
  • [PROMPT_INJECTION]: The skill provides structured templates and multi-phase instructions to guide the AI's analysis. While it ingests untrusted code, it lacks capabilities for network exfiltration or shell command injection that would make indirect prompt injection a critical risk. It includes explicit rules for reporting security vulnerabilities (like hardcoded secrets) within the analyzed codebase as a feature for the developer.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 20, 2026, 08:00 AM
Security Audit — agent-trust-hub — aidlc-reverse-engineer