Generate and Run

Generation is the step that turns your Typeweaver spec into runnable code. In the normal workflow, you update the spec, regenerate clients and server artifacts, then use that generated code from your app or runtime.

Run the generator

For a todo spec at ./api/spec/index.ts:

npx typeweaver generate --input ./api/spec/index.ts --output ./api/generated --plugins clients,hono

If you want the dependency-free server output instead of Hono:

npx typeweaver generate --input ./api/spec/index.ts --output ./api/generated --plugins clients,server

How to choose plugins

clients generates contract-aware clients and request commands.
hono generates Hono integration artifacts such as <ResourceName>Hono.
server generates dependency-free server artifacts such as <ResourceName>Router and TypeweaverApp.

You can combine plugins in one run as long as they fit your setup.

What the generated folder contains

The exact files depend on your plugins, but a generated folder usually includes:

a folder per resource, such as todo/
request and response types
request and response validators
generated clients when clients is enabled
generated server or Hono integration files when those plugins are enabled

In practice, your application code imports from ./api/generated and treats it as generated source.

When to regenerate

Regenerate whenever the spec changes.

That includes changes to:

operation paths or methods
request schemas
response schemas
resource names or operationId values
enabled plugins or output location

If the contract changed, regenerate before running the app, tests, or committing generated output.

Using a config file

If you prefer a config file, use JavaScript: .js, .mjs, or .cjs. TypeScript config files are not supported by the published CLI.

// typeweaver.config.mjs
export default {
  input: "./api/spec/index.ts",
  output: "./api/generated",
  plugins: ["clients", "server"],
};

npx typeweaver generate --config ./typeweaver.config.mjs

See CLI Reference for the full config surface.