The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill facilitates downloading the awesome-ionic-mcp server from the NPM registry and fetching documentation from various external sources.
Evidence: The setup instructions utilize npx to download and run the MCP server.
Evidence: Documentation is retrieved from external domains such as ionicframework.com, capacitorjs.com, and capgo.app.
[COMMAND_EXECUTION]: Provides tools for an AI agent to execute local CLI commands for project management, builds, and platform synchronization.
Evidence: Tools like ionic_build, capacitor_sync, capacitor_run, and ionic_generate act as wrappers for system shell commands.
Evidence: The ionic_config_set tool allows modification of project-level configuration.
[REMOTE_CODE_EXECUTION]: Utilizes the npx utility to execute remote packages at runtime.
Evidence: Configuration examples for Claude Desktop, Cline, and Cursor show execution of awesome-ionic-mcp via the npx command.
[PROMPT_INJECTION]: The skill exhibits an attack surface for indirect prompt injection because it ingests data from external websites and passes it to an agent with significant system capabilities.
Ingestion points: Data is ingested via get_component_api, get_official_plugin_api, and get_plugin_api from external web documentation.
Boundary markers: No specific delimiters or instructions to ignore embedded commands are defined.
Capability inventory: The skill exposes high-impact CLI commands including ionic_start, ionic_build, capacitor_run, and capacitor_add.
Sanitization: There is no documented mechanism for sanitizing or escaping the content retrieved from external sources before it is processed by the agent.

capacitor-mcp