develop-clean-architecture

Pass

Audited by Gen Agent Trust Hub on Apr 21, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: No malicious patterns, obfuscation, or unauthorized network operations were detected across the 81 analyzed files.
  • [COMMAND_EXECUTION]: The skill instructs the agent to perform code verification using project-native commands like npm run typecheck, npm test, and tsc --noEmit. These are standard development workflows for auditing and refactoring TypeScript codebases.
  • [INDIRECT_PROMPT_INJECTION]: The skill possesses a potential attack surface for indirect prompt injection because it processes external source code provided by the user.
  • Ingestion points: Files from the user's project (tsconfig.json, source files, project structure) are read into the agent's context during the 'Reviewing' and 'Implementing' phases.
  • Boundary markers: The skill defines explicit 'Guardrails' in SKILL.md to constrain agent behavior, though it does not use specific delimiters for user-provided code.
  • Capability inventory: The agent is authorized to write code, run shell commands for testing (npm test, tsc), and analyze project structure.
  • Sanitization: There is no explicit sanitization of the input code mentioned, relying on the LLM's inherent safety filters and the architectural constraints provided in the skill.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 21, 2026, 04:42 PM