skills/g1joshi/agent-skills/wordpress/Gen Agent Trust Hub

wordpress

Pass

Audited by Gen Agent Trust Hub on Oct 11, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill provides templates for building WordPress components, such as Gutenberg blocks and REST API endpoints, which are designed to ingest and process user-provided data. This creates a potential surface for indirect prompt injection within the developed applications if proper sanitization or boundary markers are omitted.
  • Ingestion points: The RichText component in src/index.js (user-input in block editor) and the register_rest_route callback in the PHP examples.
  • Boundary markers: Not explicitly shown in the simplified code snippets, relying on standard WordPress framework protections.
  • Capability inventory: The skill demonstrates usage of $wpdb for database access and register_rest_route for exposing API endpoints in SKILL.md.
  • Sanitization: While omitted in the quick-start snippets for brevity, the "Best Practices" section explicitly mandates the use of sanitize_text_field(), esc_html(), and esc_url() to mitigate injection risks.
  • [SAFE]: The skill serves as a legitimate technical guide for WordPress development. It follows modern standards (PHP 8.2+, Full Site Editing) and focuses on teaching secure coding patterns, such as using $wpdb->prepare() to prevent SQL injection and requiring a permission_callback for API security. All referenced domains (wordpress.org) are official and trusted repositories for WordPress documentation.
Audit Metadata
Risk Level
SAFE
Analyzed
Oct 11, 2026, 03:37 AM