architecture-decision-records

Pass

Audited by Gen Agent Trust Hub on Apr 6, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill's primary purpose is to automate the creation and maintenance of Architecture Decision Records (ADRs) as Markdown files. It does not contain any executable code, scripts, or external dependencies.
  • [COMMAND_EXECUTION]: The skill performs file system operations (creating directories and writing files). However, the workflow strictly requires explicit user confirmation before creating the docs/adr/ directory or writing any new ADR files, which prevents unauthorized file system modifications.
  • [DATA_EXFILTRATION]: There are no network operations, API calls, or data transmission patterns. The skill's scope is limited to the local project's documentation directory.
  • [INDIRECT_PROMPT_INJECTION]: While the skill reads existing ADR files from the project directory, it uses this information solely to inform the user about previous decisions. It does not pass this content into sensitive system commands or use it to bypass safety constraints.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 6, 2026, 04:00 AM
Security Audit — agent-trust-hub — architecture-decision-records