emdash-cms

Pass

Audited by Gen Agent Trust Hub on Sep 14, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill provides functionality to ingest external content and integrate with AI tools, creating a surface for indirect prompt injection.
  • Ingestion points: The skill supports importing content from external WordPress XML export files and REST APIs via the CLI (SKILL.md), and provides a built-in MCP server for AI tool access to CMS data (SKILL.md).
  • Boundary markers: The MCP server implementation includes configuration for role-based access (allowedRoles) and requires an authorization token for connection (SKILL.md).
  • Capability inventory: The plugin system grants capabilities for network operations (http:fetch), email transmission (email:send), and persistent storage writes (kv:write, storage:write) as defined in the manifest (SKILL.md).
  • Sanitization: External content from WordPress is processed through the gutenberg-to-portable-text converter, which transforms raw data into a structured Portable Text JSON format (SKILL.md).
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 14, 2026, 10:36 AM