wordpress-router
Pass
Audited by Gen Agent Trust Hub on Sep 14, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill analyzes external repository files (such as composer.json, package.json, style.css, and block.json) via a triage script to classify the project type. This represents a potential surface for indirect prompt injection where malicious content in these files could influence the agent's routing decisions.
- Ingestion points: Project root files read by detect_wp_project.mjs and manual inspection of configuration files.
- Boundary markers: The instructions do not specify explicit delimiters or warnings for the agent to ignore instructions embedded within the analyzed project files.
- Capability inventory: The skill facilitates execution of build tools, test runners, and domain-specific WordPress skills based on the classification.
- Sanitization: Classification relies on keyword matching and structured data property checks which provides some natural validation against arbitrary text injection.
Audit Metadata