The Agent Skills Directory

[COMMAND_EXECUTION]: The skill instructs the agent to execute shell commands using bun, bunx, npm, yarn, and pnpm to install dependencies and run the @hey-api/openapi-ts CLI for code generation.
[EXTERNAL_DOWNLOADS]: The skill facilitates downloading OpenAPI specifications from remote URLs and official registries (e.g., Scalar, ReadMe). It also references documentation from https://heyapi.dev and GitHub.
[PROMPT_INJECTION]: The skill exhibits surface area for indirect prompt injection as it processes untrusted OpenAPI specifications to generate source code.
Ingestion points: Remote or local OpenAPI specification files provided via the input configuration field in setup-config.md.
Boundary markers: None explicitly instructed for the generated output files.
Capability inventory: The skill performs file system writes (generating code) and shell command execution (running the generator).
Sanitization: The skill relies on the core @hey-api/openapi-ts library's parser to handle input specifications.
[CREDENTIALS_UNSAFE]: The documentation includes a note on using NODE_TLS_REJECT_UNAUTHORIZED=0 to bypass certificate validation for self-signed certificates during development, which is a common but insecure practice if applied to production environments.

heyapi