throw

neuro.iterator.throw

Instance method on Iterator.prototype.

Signatures

throw(input: { iterator: <receiver>; e?: any; prompt?: string }): Promise<IteratorResult<T, TReturn>>

The prompt field is optional. When omitted (or set to an empty string) the wrapper falls back to the native Iterator.prototype.throw and returns a resolved Promise without contacting the LLM. When present, the LLM is given the original arguments plus your prompt and is asked to behave like the original method.

Example

import { configureClient, neuro } from 'neuro-ts';

configureClient({ apiKey: process.env.OPENAI_API_KEY });

// Inject error into iterator; rarely caught gracefully, almost always rethrown.
await neuro.iterator.throw({ iterator: it, e: new Error('cancel'), prompt: 'inject an exception into the iterator at the next yield point -- the protocol method nobody calls directly because every test that exercised it discovered the generator did not handle it' });

System prompt

The exact system prompt the SDK sends to your model when you provide a prompt field:

Generated promptIterator.prototype.throw

You are simulating the JavaScript built-in `Iterator.prototype.throw`.
## Original signature(s)
  Overload 1: (e?: any) => IteratorResult<T, TReturn>
## How to respond
- Behave EXACTLY as the original `throw` would, but use the user's intent to choose any callback / comparator / transform logic that the original would normally accept as an argument.
- Strictly preserve the original return type and shape.
- Output ONLY the JSON-encoded return value of the function call.
- Do NOT include explanations, prose, comments, or markdown fences.
- If the function would return `undefined`, output the literal string `undefined`.
- For Date / RegExp / Map / Set / TypedArray returns, output an object of the form { "__type": "Date" | "RegExp" | "Map" | "Set" | "<TypedArrayName>", ... } so the SDK can rehydrate it.