privacy-review-rails

Pass

Audited by Gen Agent Trust Hub on Sep 22, 2026

Risk Level: SAFECOMMAND_EXECUTIONDYNAMIC_CONTEXT_INJECTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [DYNAMIC_CONTEXT_INJECTION]: The skill uses dynamic context injection via the !command syntax to verify the presence of the privacy-by-design-rails dependency on the local file system during skill initialization.
  • [COMMAND_EXECUTION]: The skill invokes a local Ruby scanner script using the Bash tool to analyze the application's code for privacy violations.
  • [INDIRECT_PROMPT_INJECTION]: The skill analyzes untrusted code from the Rails project (e.g., models, controllers, mailers), creating a surface for potential prompt injection from malicious code content.
  • Ingestion points: Rails application source files (models, migrations, mailers, jobs, controllers, initializers, services).
  • Boundary markers: None; the skill reads and processes the full content of these files without specific isolation or instructions to ignore embedded commands.
  • Capability inventory: The skill can execute shell commands (Bash), read files (Read, Glob, Grep), and modify code (Edit).
  • Sanitization: The skill does not perform sanitization or validation of the code content before analysis.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 22, 2026, 03:25 AM
Security Audit — agent-trust-hub — privacy-review-rails