codebase-design

Pass

Audited by Gen Agent Trust Hub on Oct 8, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill consists of documentation and procedural instructions for software engineering design patterns. It does not contain any executable code, scripts, or network-enabled operations.
  • [SAFE]: External references are limited to a well-known open-source repository on GitHub (mattpocock/skills) for attribution purposes, which is consistent with the skill's stated origin.
  • [INDIRECT_PROMPT_INJECTION]: The DESIGN-IT-TWICE.md file describes a workflow involving spawning sub-agents based on user-supplied technical briefs. While this represents a data ingestion surface, the skill is scoped to design analysis and does not grant the agent high-privilege capabilities such as file system writes or arbitrary command execution.
  • [DYNAMIC_EXECUTION]: The skill instructs the agent to use an 'Agent tool' to orchestrate sub-agents for parallel design exploration. This is a standard architectural pattern for complex reasoning tasks and does not involve the execution of untrusted code snippets.
Audit Metadata
Risk Level
SAFE
Analyzed
Oct 8, 2026, 06:31 PM