microservices-patterns

Pass

Audited by Gen Agent Trust Hub on Apr 6, 2026

Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
  • [PROMPT_INJECTION]: The skill processes untrusted user-provided source code in its 'Code Review' mode, creating a potential surface for indirect prompt injection where malicious instructions could be embedded in the code to influence the agent's behavior or output.
  • Ingestion points: User-supplied code for review via Mode 2 in SKILL.md.
  • Boundary markers: Absent; the instructions do not mandate the use of delimiters or 'ignore embedded instructions' prompts when analyzing external code.
  • Capability inventory: The skill can generate complex architectural plans and invokes a Python script that performs file system writes.
  • Sanitization: Absent; there is no specified validation or filtering of the input code before processing.
  • [COMMAND_EXECUTION]: The skill includes a Python utility script 'scripts/new_service.py' used to generate microservice scaffolds.
  • The script creates directory structures and writes multiple source files (Python, Java, or Kotlin) to the local filesystem.
  • It uses string templates for safe variable substitution and does not execute the generated code directly, though it provides instructions for the user to do so.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 6, 2026, 11:07 AM
Security Audit — agent-trust-hub — microservices-patterns