ionic-expert

Pass

Audited by Gen Agent Trust Hub on Sep 15, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill provides instructions for global package installation via npm install -g @ionic/cli and local project scaffolding using ionic start. It also lists various development commands such as ionic serve, ionic build, and ionic cap run, including the --external flag which binds the development server to all network interfaces.
  • [EXTERNAL_DOWNLOADS]: The instructions include adding two hosted Model Context Protocol (MCP) servers using the claude mcp add command targeting mcp.capawesome.io and ionic-framework-mcp.capawesome.io. These are recognized as vendor-managed infrastructure for providing live documentation updates.
  • [INDIRECT_PROMPT_INJECTION]: The skill directs the agent to query external MCP servers for documentation updates using search_docs and get_doc_page tools, which introduces a surface for indirect injection.
  • Ingestion points: Data returned from external MCP tool calls (search_docs, get_doc_page) targeting remote documentation servers.
  • Boundary markers: The skill does not define specific delimiters or instructions for the agent to disregard instructions embedded within the fetched documentation content.
  • Capability inventory: The skill environment allows for shell command execution (npm, ionic, claude), project file modification, and network communication via the Ionic CLI and MCP transport.
  • Sanitization: There is no evidence of sanitization, validation, or schema enforcement for the content retrieved from the external documentation servers.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 15, 2026, 08:00 AM