chore: generate

packages/llm/DESIGN.md

@@ -242,7 +242,8 @@ executable tools. Call-level values override model defaults.
 Provider-specific options are inferred from the concrete model:
 
 ```ts
-yield* LLM.generate({
+yield *
+  LLM.generate({
     model: OpenAI.model("gpt-4.1-mini"),
     prompt: "Hello",
     provider: {
@@ -278,7 +279,9 @@ provider.
 ### Inline input
 
 ```ts
-const result = yield* LLM.generate({
+const result =
+  yield *
+  LLM.generate({
     model,
     system: "You are concise.",
     prompt: "Summarize this pull request.",
@@ -362,7 +365,9 @@ const tools = {
   }),
 }
 
-const result = yield* LLM.generate({
+const result =
+  yield *
+  LLM.generate({
     model,
     prompt: "What is the weather in London?",
     tools,
@@ -387,14 +392,13 @@ successful result with `stopReason: "max-turns"`, not an Effect failure.
 ### Custom stopping
 
 ```ts
-const result = yield* LLM.generate({
+const result =
+  yield *
+  LLM.generate({
     model,
     prompt,
     tools,
-  stopWhen: StopWhen.any(
+    stopWhen: StopWhen.any(StopWhen.turnCount(8), StopWhen.hasToolCall("finalize")),
-    StopWhen.turnCount(8),
-    StopWhen.hasToolCall("finalize"),
-  ),
   })
 ```
 
@@ -433,11 +437,7 @@ const call = Array.from(events).find(LLMEvent.is.toolCall)
 if (call && !call.providerExecuted) {
   const dispatched = yield * ToolRuntime.dispatch(tools, call)
   const followUp = LLM.updateRequest(request, {
-    messages: [
+    messages: [...request.messages, Message.assistant([call]), Message.tool({ ...call, result: dispatched.result })],
-      ...request.messages,
-      Message.assistant([call]),
-      Message.tool({ ...call, result: dispatched.result }),
-    ],
   })
   // Caller must invoke the provider again and repeat the loop.
 }
@@ -452,7 +452,9 @@ OpenCode and other durable runtimes need to own persistence, tool settlement,
 and continuation. They use the explicit turn API:
 
 ```ts
-const result = yield* LLM.generateTurn({
+const result =
+  yield *
+  LLM.generateTurn({
     model,
     request,
     // Definitions only. generateTurn never dispatches local handlers.
@@ -494,7 +496,9 @@ const request = LLM.request({
   },
 })
 
-const result = yield* LLM.generate({
+const result =
+  yield *
+  LLM.generate({
     model,
     request,
     tools: {
@@ -516,7 +520,9 @@ binding. Missing or incompatible bindings fail with a typed tool-binding error.
 Provider-hosted tools are distinct typed values:
 
 ```ts
-const result = yield* LLM.generate({
+const result =
+  yield *
+  LLM.generate({
     model: OpenAI.model("gpt-4.1"),
     prompt: "Find today's relevant announcements.",
     tools: {
@@ -589,7 +595,9 @@ const Weather = Schema.Struct({
   highCelsius: Schema.Number,
 })
 
-const result = yield* LLM.generate({
+const result =
+  yield *
+  LLM.generate({
     model,
     prompt: "Give me today's weather for London.",
     output: Weather,
@@ -612,7 +620,9 @@ Advanced callers may override the strategy when exact provider semantics matter.
 
 ```ts
 // Current API is a separate operation and always forces a synthetic tool.
-const result = yield* LLM.generateObject({
+const result =
+  yield *
+  LLM.generateObject({
     model,
     prompt,
     schema: Weather,
@@ -679,7 +689,8 @@ cache boundaries where explicit caching is supported and does nothing on the wir
 where providers cache implicitly.
 
 ```ts
-yield* LLM.generate({
+yield *
+  LLM.generate({
     model,
     prompt,
     cache: "none", // Explicit opt-out.
@@ -705,7 +716,8 @@ silently inherit custom retry policies.
 ### Timeouts
 
 ```ts
-yield* LLM.generate({
+yield *
+  LLM.generate({
     model,
     prompt,
     timeout: "2 minutes", // Entire run, including tools.
@@ -740,7 +752,8 @@ retry, or redirect control flow.
 ```ts
 const model = OpenAI.model("gpt-4.1", {
   hooks: {
-    request: (request) => Effect.succeed({
+    request: (request) =>
+      Effect.succeed({
         ...request,
         metadata: { ...request.metadata, tenant: "acme" },
       }),
@@ -775,7 +788,8 @@ The request customization ladder is:
 5. Experimental provider-definition or protocol patching
 
 ```ts
-yield* LLM.generate({
+yield *
+  LLM.generate({
     model,
     prompt,
     http: {
@@ -928,10 +942,7 @@ Provider authoring is public but experimental.
 ### Declarative provider definition
 
 ```ts
-import {
+import { Provider, Protocol } from "@opencode-ai/ai/provider"
-  Provider,
-  Protocol,
-} from "@opencode-ai/ai/provider"
 
 export const ExampleAI = Provider.define({
   id: "example",
@@ -996,9 +1007,7 @@ SDK integrations that motivated this package.
 const PatchedResponses = OpenAIResponses.with({
   body: {
     fromRequest: (request) =>
-      OpenAIResponses.body.fromRequest(request).pipe(
+      OpenAIResponses.body.fromRequest(request).pipe(Effect.map((body) => ({ ...body, custom_field: true }))),
-        Effect.map((body) => ({ ...body, custom_field: true })),
-      ),
   },
   stream: {
     step: patchResponsesStep,
@@ -1043,7 +1052,7 @@ providers, and there is no preferred all-providers barrel.
 ## Defaults
 
 | Concern                      | Default                                             |
-| --- | --- |
+| ---------------------------- | --------------------------------------------------- |
 | `LLM.generate` semantics     | Complete Model Run                                  |
 | `LLM.generateTurn` semantics | Exactly one Provider Turn                           |
 | Maximum turns                | 20                                                  |
@@ -1064,7 +1073,7 @@ providers, and there is no preferred all-providers barrel.
 The redesign intentionally removes or changes these current concepts:
 
 | Current                                 | Proposed                                                    |
-| --- | --- |
+| --------------------------------------- | ----------------------------------------------------------- |
 | `@opencode-ai/llm`                      | `@opencode-ai/ai`                                           |
 | Mandatory `LLM.request({ model, ... })` | Inline calls or model-free portable requests                |
 | `LLM.generate` means one turn           | `LLM.generate` means complete run                           |