isolating-product-facade-contracts

Pass

Audited by Gen Agent Trust Hub on Jun 24, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: No malicious patterns, obfuscation, or unauthorized data access were detected. All described actions are consistent with legitimate software engineering practices for architectural refactoring within a controlled monorepo environment.
  • [COMMAND_EXECUTION]: The skill instructs the agent to use local CLI tools, including hogli (a PostHog-specific utility) and tach (an architecture enforcement tool). These tools are used for scanning the codebase, linting imports, and moving files to their new architectural locations. These operations are localized and standard for the described engineering task.
  • [PROMPT_INJECTION]: The skill does not contain any attempts to override system prompts or bypass safety guidelines. Its instructions are purely technical and focused on code migration workflows.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 24, 2026, 06:17 PM
Security Audit — agent-trust-hub — isolating-product-facade-contracts