lw-site-manager-overview

Pass

Audited by Gen Agent Trust Hub on Sep 9, 2026

Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill provides documentation and examples for an AI agent to execute a wide array of administrative functions on a WordPress site. These include managing plugins, themes, users, and database operations. While these actions are structured through an API, they represent significant functional control over the target environment.
  • [INDIRECT_PROMPT_INJECTION]:
  • Ingestion points: The skill describes a REST API and PHP execution pattern that accepts structured input for over 120 administrative abilities (found in SKILL.md).
  • Boundary markers: The provided documentation does not include explicit instructions for the agent to ignore or delimit potentially malicious instructions embedded within the input data.
  • Capability inventory: The skill facilitates high-privilege operations such as site backups, database repairs, and user role management (found in SKILL.md).
  • Sanitization: The skill references the use of a centralized PermissionManager for WordPress capability checks and JSON Schema validation for input and output (found in SKILL.md).
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 9, 2026, 07:07 PM
Security Audit — agent-trust-hub — lw-site-manager-overview