The Agent Skills Directory

[SAFE]: The skill serves its intended purpose of providing technical documentation for the oRPC v2 framework. All code snippets and patterns align with standard software development practices.
[EXTERNAL_DOWNLOADS]: The skill references documentation resources at v2.orpc.dev. These are official project domains used for providing context to the AI agent.
[COMMAND_EXECUTION]: The skill includes common development CLI commands such as npx @hey-api/openapi-ts for code generation and tsdown for bundling. These are routine operations for the described workflow.
[INDIRECT_PROMPT_INJECTION]: The skill defines a workflow for ingesting external OpenAPI specifications to generate code, which is an expected attack surface for API development tools.
Ingestion points: The configuration for Hey API in SKILL.md allows specifying URLs or local files as input.
Boundary markers: No specific delimiters are defined in the instructions for the ingested schema data.
Capability inventory: The skill utilizes local file system writes (fs.writeFileSync) and code generation tools (@hey-api/openapi-ts).
Sanitization: The skill relies on standard third-party tools to process and validate external inputs.

orpc-contract