The Agent Skills Directory

[COMMAND_EXECUTION]: The skill performs several automated shell operations to manage the migration workflow. This includes git commands for branch creation and staging changes, as well as grep and sed utilities used to safely extract environment variable names from local .env files without accessing their values.
[EXTERNAL_DOWNLOADS]: The skill uses npx --yes @adobe/aio-commerce-lib-app@latest init to download and execute the official Adobe Commerce App Builder CLI tool directly from the npm registry. This is a standard initialization procedure for the framework.
[INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data from the local project, which serves as a potential attack surface for indirect prompt injection.
Ingestion points: Project metadata from package.json (specifically the description field) and content from README.md are read by the Analyzer agent.
Boundary markers: The skill does not employ specific delimiters or boundary markers when processing or rewriting these strings for the new configuration file.
Capability inventory: The skill has the ability to write files to the filesystem, execute CLI tools via shell, and perform git operations.
Sanitization: While the skill performs length-based rewriting of descriptions to fit schema limits, it does not implement specific sanitization to filter potential injection patterns from the ingested text.
[DYNAMIC_EXECUTION]: The Executor agent generates new TypeScript configuration files (app.commerce.config.ts) and scaffolds JavaScript action handlers in the src/ directory by merging data extracted during the analysis phase with predefined templates.

commerce-app-migrate