software-architecture

Pass

Audited by Gen Agent Trust Hub on Sep 5, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill instructs the agent to "ALWAYS search for existing solutions before writing custom code" and specifically to "Check npm for existing libraries". This establishes a vulnerability surface where the agent may fetch and integrate content from external, third-party sources that could contain malicious instructions or code.
  • Ingestion points: External package registries (e.g., NPM), third-party APIs, and SaaS service documentation.
  • Boundary markers: The instructions lack explicit boundary markers or warnings to the agent to ignore embedded instructions within these external sources.
  • Capability inventory: The skill is designed for software development, which typically involves the ability to write, analyze, and potentially execute code or install packages.
  • Sanitization: There are no instructions or mechanisms provided to sanitize or validate the integrity of the libraries or documentation retrieved from the web.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 5, 2026, 02:02 PM
Security Audit — agent-trust-hub — software-architecture