software-architecture

Pass

Audited by Gen Agent Trust Hub on Aug 22, 2026

Risk Level: SAFENO_CODE
Full Analysis
  • [SAFE]: The skill is composed entirely of instructional markdown files and architectural templates. It contains no executable scripts, binaries, or runtime code that could be used for malicious purposes.
  • [SAFE]: All external references provided in the source index point to reputable, well-known industry standards and documentation from organizations such as AWS, Google, NIST, and ISO. These are provided for informational context and are not involved in automated downloads or executions.
  • [SAFE]: The skill clearly defines 'Ownership Boundaries', explicitly instructing the agent to delegate sensitive or implementation-specific tasks—such as security engineering, platform provisioning, and data platform design—to other specialized skills, adhering to the principle of least privilege.
  • [SAFE]: No patterns of prompt injection, data exfiltration, or obfuscation were detected. The instructional content is professional and focused exclusively on system design methodology.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 22, 2026, 03:44 AM
Security Audit — agent-trust-hub — software-architecture