adr-writing-weapon

Pass

Audited by Gen Agent Trust Hub on Jul 25, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill's primary function is to assist users in authoring and maintaining Architecture Decision Records (ADRs) using established templates like Nygard, MADR, and Y-statements. All content analyzed consists of benign documentation, guides, and templates.
  • [EXTERNAL_DOWNLOADS]: The skill references and provides installation instructions for well-known open-source tools such as adr-tools and log4brains via official package managers (NPM, Homebrew). These are standard tools in the software architecture community and do not represent a security risk.
  • [DATA_EXFILTRATION]: No indicators of unauthorized data access, sensitive file reads, or network exfiltration were detected. The skill operates on local documentation files.
  • [PROMPT_INJECTION]: The instructions do not contain any patterns aimed at bypassing AI safety protocols, extracting system prompts, or overriding agent constraints. The 'Do NOT use for' section is a standard scope definition for agent task-shifting.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 25, 2026, 03:22 AM
Security Audit — agent-trust-hub — adr-writing-weapon