php-ddd-architecture

Pass

Audited by Gen Agent Trust Hub on Sep 19, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill instructs the agent to ingest and follow instructions from project-local files such as AGENTS.md, CLAUDE.md, and domain documentation. This creates a surface where malicious instructions embedded in these files could influence the agent's behavior.
  • Ingestion points: AGENTS.md, CLAUDE.md, and project-specific domain documentation.
  • Boundary markers: The skill includes an "Activation Guard" requiring an explicit DDD declaration, but does not provide delimiters or instructions to ignore embedded commands within the processed data.
  • Capability inventory: The skill enables the agent to perform comprehensive architectural changes, including creating/modifying entities, repositories, and framework configurations.
  • Sanitization: No sanitization or validation mechanisms are defined for the instructions read from project files.
  • [SAFE]: The guidelines for repository contracts, framework configuration, and authentication boundaries align with industry best practices for Domain-Driven Design in PHP environments (specifically using Slim, Mezzio, and Laminas). The repository examples for file and API access are standard architectural patterns.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 19, 2026, 10:34 AM
Security Audit — agent-trust-hub — php-ddd-architecture