The Agent Skills Directory

[DYNAMIC_CONTEXT_INJECTION]: The skill utilizes platform-specific syntax to execute shell commands when the skill is loaded. It runs a Node.js script to extract information from package.json and a find command to locate configuration files (config.xml, capacitor.config.json, etc.). This data is used to populate a 'Live Project Snapshot' that informs the agent about the current state of the application's mobile framework dependencies.
[INDIRECT_PROMPT_INJECTION]: The skill ingests data from the local filesystem during the dynamic context injection phase. Ingestion points: Project configuration and dependency files (package.json) are read and output to the prompt context via shell commands. Boundary markers: The injected data is not enclosed in specific delimiters. Capability inventory: The skill is configured to allow shell execution via Bash for analysis and migration tasks, and it explicitly guides the agent to install external packages using npm and npx. Sanitization: No sanitization or filtering is applied to the values extracted from package.json before they are presented to the agent.
[COMMAND_EXECUTION]: The migration instructions guide the agent to perform numerous system operations, including installing packages from the npm registry and managing native platforms via CLI tools. These actions are standard for the described purpose of the skill.

cordova-to-capacitor