architectural-decisions

Pass

Audited by Gen Agent Trust Hub on Nov 1, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill facilitates architectural decision tracking (ADRs) within a codebase using the repowise tool. It provides standard instructional guidance for querying design rationale and recording new decisions.
  • [COMMAND_EXECUTION]: The skill instructs the agent to use the repowise CLI tool (e.g., repowise decision add, repowise decision candidates). These are project-specific developer tools from the author ('repowise-dev') used as intended for architectural management.
  • [INDIRECT_PROMPT_INJECTION]: While the skill processes data from commit messages and code comments (e.g., markers like WHY:, DECISION:), this represents standard static analysis of codebase metadata. There is no evidence of unsafe interpolation or automated execution of untrusted instructions from these sources.
Audit Metadata
Risk Level
SAFE
Analyzed
Nov 1, 2026, 04:06 PM