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

wpds

Pass

Audited by Gen Agent Trust Hub on Sep 14, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTIONDYNAMIC_EXECUTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill is configured to ingest external content from a WordPress Design System (WPDS) MCP server using the wpds:// URI scheme. While this creates a surface for indirect prompt injection from the retrieved documentation or component data, the instructions prioritize these sources as canonical for design tasks.
  • Ingestion points: External data retrieved via wpds://pages, wpds://components, and wpds://design-tokens (SKILL.md).
  • Boundary markers: The skill instructs the agent to treat these sources as authoritative and push back against conflicting web-searched information.
  • Capability inventory: No file-writing or direct network operations are defined within the skill's own scripts.
  • Sanitization: Not explicitly defined for the ingested MCP data.
  • [DYNAMIC_EXECUTION]: The skill encourages the use of lint scripts provided by the user's local context to validate code output. This is a standard developer practice and does not involve the skill downloading or executing unverified external scripts.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 14, 2026, 07:58 PM
Security Audit — agent-trust-hub — wpds