craft-site
Pass
Audited by Gen Agent Trust Hub on Oct 28, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: No malicious patterns, obfuscation, or persistence mechanisms were detected. The skill instructions align with standard web development practices and focus on front-end template architecture.
- [INDIRECT_PROMPT_INJECTION]: The skill documents patterns for processing user-supplied data and external content. Ingestion points: includes craft.app.request (GET/POST), entry content, and external feeds. Boundary markers: Explicit escaping rules, CDATA blocks for XML, and CSRF tokens are consistently recommended. Capability inventory: CLI operations via DDEV and network requests for documentation or API integrations. Sanitization: Detailed best practices are provided, including mandatory |e filter usage, striptags, CSRF validation, and same-site redirect checks to mitigate potential injection risks.
- [EXTERNAL_DOWNLOADS]: The skill documents the integration of well-known third-party services such as Google Tag Manager, UserCentrics, CookieBot, and Fathom Analytics. These are industry-standard tools and their inclusion is documented neutrally for development purposes.
- [CREDENTIALS_UNSAFE]: The documentation promotes safe credential handling by using environment variable lookups (App::env) for sensitive keys. All tokens used in examples (e.g., in GraphQL documentation) are generic placeholders for illustration.
Audit Metadata