authoring-architecture-doc

Pass

Audited by Gen Agent Trust Hub on Jun 14, 2026

Risk Level: SAFEPROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection as it ingests untrusted external documents (PRDs and product direction) to generate architecture documentation.
  • Ingestion points: The skill explicitly instructs the agent to read all documents in the 'depends_on' set, specifically naming the PRD and product direction files in SKILL.md.
  • Boundary markers: The instructions do not specify the use of delimiters or 'ignore instructions' warnings for the processed data.
  • Capability inventory: The skill uses the input to generate whole-system architecture documentation, which influences structural decisions, technology selection, and security mechanism realizations.
  • Sanitization: There is no workflow defined for validating or filtering the content of the provided documents.
  • [EXTERNAL_DOWNLOADS]: The skill contains references to external documentation and repositories in references/sources.md to provide research provenance for its methodology.
  • Details: References include well-known technological resources such as arc42.org, c4model.com, iso-architecture.org, and various ADR-related repositories. These are documented as established industry practices and standards.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 14, 2026, 07:32 PM
Security Audit — agent-trust-hub — authoring-architecture-doc