The Agent Skills Directory

[COMMAND_EXECUTION]: The skill executes development-related commands using the Nango CLI and npm.
Evidence: Instructions include running nango compile, nango dev, nango dryrun, nango generate:tests, and npm test to validate the migration.
[INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted data from existing project files to generate code, which constitutes an indirect prompt injection surface.
Ingestion points: Reads nango.yaml and existing TypeScript source files from the repository root.
Boundary markers: No explicit boundary markers are defined for the input data, but the skill provides strict mapping rules for conversion.
Capability inventory: Writing multiple project files (models.ts, index.ts, package.json, etc.) and executing shell commands via the CLI.
Sanitization: The skill provides a structured schema conversion logic (Zod-based) which acts as a validation layer for the transformed data.
[DYNAMIC_EXECUTION]: The skill generates new TypeScript files and configuration at runtime, which are then compiled and executed.
Evidence: Instructions guide the creation of models.ts and the rewriting of sync/action files using boilerplate templates, followed by execution through the nango compile and dryrun commands.

migrating-to-zero-yaml