architecture-decision

Pass

Audited by Gen Agent Trust Hub on May 4, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill's operations are consistent with its documented purpose of managing Architectural Decision Records (ADRs).
  • [PROMPT_INJECTION]: The skill possesses a surface for indirect prompt injection because it reads and processes external documents (GDDs, engine references) to generate content. This is a functional requirement of the skill and risk is managed through human-in-the-loop checkpoints.
  • Ingestion points: design/gdd/, docs/engine-reference/, docs/architecture/ (SKILL.md)
  • Boundary markers: Implicit in collaborative design steps
  • Capability inventory: Write, Edit tools (SKILL.md)
  • Sanitization: Data is extracted and summarized rather than directly executed
  • [COMMAND_EXECUTION]: The skill uses the Task tool to orchestrate reviews by specialized sub-agents. These operations are governed by local documentation and do not represent arbitrary command execution.
Audit Metadata
Risk Level
SAFE
Analyzed
May 4, 2026, 12:49 PM