@miaws/miaw 1.18.4 → 1.18.5

package/src/session/llm/native-request.ts

@@ -0,0 +1,196 @@
+import type { JsonSchema, LLMRequest, ProviderMetadata } from "@miaw/llm"
+import { LLM, Message, SystemPart, ToolCallPart, ToolDefinition, ToolResultPart } from "@miaw/llm"
+import {
+  AmazonBedrock,
+  Anthropic,
+  Azure,
+  Google,
+  OpenAI,
+  OpenAICompatible,
+  OpenRouter,
+} from "@miaw/llm/providers"
+import type { ModelMessage } from "ai"
+import type { Provider } from "@/provider/provider"
+import { isRecord } from "@/util/record"
+
+type ToolInput = {
+  readonly description?: string
+  readonly inputSchema?: unknown
+}
+
+export type RequestInput = {
+  readonly model: Provider.Model
+  readonly apiKey?: string
+  readonly baseURL?: string
+  readonly system?: readonly string[]
+  readonly messages: readonly ModelMessage[]
+  readonly tools?: Record<string, ToolInput>
+  readonly toolChoice?: "auto" | "required" | "none"
+  readonly temperature?: number
+  readonly topP?: number
+  readonly topK?: number
+  readonly maxOutputTokens?: number
+  readonly providerOptions?: LLMRequest["providerOptions"]
+  readonly headers?: Record<string, string>
+}
+
+const providerMetadata = (value: unknown): ProviderMetadata | undefined => {
+  if (!isRecord(value)) return undefined
+  const result = Object.fromEntries(
+    Object.entries(value).filter((entry): entry is [string, Record<string, unknown>] => isRecord(entry[1])),
+  )
+  return Object.keys(result).length === 0 ? undefined : result
+}
+
+// Stored AI SDK parts historically kept provider-owned continuation metadata in
+// `providerOptions`; native parts now use `providerMetadata` directly.
+const partProviderMetadata = (part: Record<string, unknown>) =>
+  providerMetadata(part.providerMetadata) ?? providerMetadata(part.providerOptions)
+
+const textPart = (part: Record<string, unknown>) => ({
+  type: "text" as const,
+  text: typeof part.text === "string" ? part.text : "",
+  providerMetadata: partProviderMetadata(part),
+})
+
+const mediaPart = (part: Record<string, unknown>) => {
+  if (typeof part.data !== "string" && !(part.data instanceof Uint8Array))
+    throw new Error("Native LLM request adapter only supports file parts with string or Uint8Array data")
+  return {
+    type: "media" as const,
+    mediaType: typeof part.mediaType === "string" ? part.mediaType : "application/octet-stream",
+    data: part.data,
+    filename: typeof part.filename === "string" ? part.filename : undefined,
+  }
+}
+
+const toolResult = (part: Record<string, unknown>) => {
+  const output = isRecord(part.output) ? part.output : { type: "json", value: part.output }
+  const type = output.type === "text" ? "text" : output.type === "error-text" ? "error" : "json"
+  return ToolResultPart.make({
+    id: typeof part.toolCallId === "string" ? part.toolCallId : "",
+    name: typeof part.toolName === "string" ? part.toolName : "",
+    result: "value" in output ? output.value : output,
+    resultType: type,
+    providerExecuted: typeof part.providerExecuted === "boolean" ? part.providerExecuted : undefined,
+    providerMetadata: partProviderMetadata(part),
+  })
+}
+
+const contentPart = (part: unknown) => {
+  if (!isRecord(part)) throw new Error("Native LLM request adapter only supports object content parts")
+  if (part.type === "text") return textPart(part)
+  if (part.type === "file") return mediaPart(part)
+  if (part.type === "reasoning")
+    return {
+      type: "reasoning" as const,
+      text: typeof part.text === "string" ? part.text : "",
+      providerMetadata: partProviderMetadata(part),
+    }
+  if (part.type === "tool-call")
+    return ToolCallPart.make({
+      id: typeof part.toolCallId === "string" ? part.toolCallId : "",
+      name: typeof part.toolName === "string" ? part.toolName : "",
+      input: part.input,
+      providerExecuted: typeof part.providerExecuted === "boolean" ? part.providerExecuted : undefined,
+      providerMetadata: partProviderMetadata(part),
+    })
+  if (part.type === "tool-result") return toolResult(part)
+  throw new Error(`Native LLM request adapter does not support ${String(part.type)} content parts`)
+}
+
+const content = (value: ModelMessage["content"]) =>
+  typeof value === "string" ? [{ type: "text" as const, text: value }] : value.map(contentPart)
+
+const messages = (input: readonly ModelMessage[]) => {
+  const system = input.flatMap((message) => (message.role === "system" ? [SystemPart.make(message.content)] : []))
+  const messages = input.flatMap((message) => {
+    if (message.role === "system") return []
+    return [
+      Message.make({
+        role: message.role,
+        content: content(message.content),
+        native: isRecord(message.providerOptions) ? { providerOptions: message.providerOptions } : undefined,
+      }),
+    ]
+  })
+  return { system, messages }
+}
+
+const schema = (value: unknown): JsonSchema => {
+  if (!isRecord(value)) return { type: "object", properties: {} }
+  if (isRecord(value.jsonSchema)) return value.jsonSchema
+  return value
+}
+
+const tools = (input: Record<string, ToolInput> | undefined): ToolDefinition[] =>
+  Object.entries(input ?? {}).map(([name, item]) =>
+    ToolDefinition.make({
+      name,
+      description: item.description ?? "",
+      inputSchema: schema(item.inputSchema),
+    }),
+  )
+
+const generation = (input: RequestInput) => {
+  const result = {
+    temperature: input.temperature,
+    topP: input.topP,
+    topK: input.topK,
+    maxTokens: input.maxOutputTokens,
+  }
+  return Object.values(result).some((value) => value !== undefined) ? result : undefined
+}
+
+const baseURL = (input: Provider.Model | RequestInput) =>
+  "model" in input ? (input.baseURL ?? (input.model.api.url || undefined)) : input.api.url || undefined
+
+const requireBaseURL = (model: Provider.Model, url: string | undefined) => {
+  if (url) return url
+  throw new Error(`Native LLM request adapter requires a base URL for ${model.providerID}/${model.id}`)
+}
+
+export const model = (input: Provider.Model | RequestInput, headers?: Record<string, string>) => {
+  const model = "model" in input ? input.model : input
+  const url = baseURL(input)
+  const options = {
+    ...("model" in input && input.apiKey ? { apiKey: input.apiKey } : {}),
+    ...(url ? { baseURL: url } : {}),
+    headers: Object.keys({ ...model.headers, ...headers }).length === 0 ? undefined : { ...model.headers, ...headers },
+    limits: {
+      context: model.limit.context,
+      output: model.limit.output,
+    },
+  }
+  if (model.api.npm === "@ai-sdk/openai") return OpenAI.configure(options).responses(model.api.id)
+  if (model.api.npm === "@ai-sdk/azure")
+    return Azure.configure({ ...options, baseURL: requireBaseURL(model, url) }).responses(model.api.id)
+  if (model.api.npm === "@ai-sdk/anthropic") return Anthropic.configure(options).model(model.api.id)
+  if (model.api.npm === "@ai-sdk/google") return Google.configure(options).model(model.api.id)
+  if (model.api.npm === "@ai-sdk/amazon-bedrock") return AmazonBedrock.configure(options).model(model.api.id)
+  if (model.api.npm === "@ai-sdk/openai-compatible")
+    return OpenAICompatible.configure({
+      ...options,
+      provider: String(model.providerID),
+      baseURL: requireBaseURL(model, url),
+    }).model(model.api.id)
+  if (model.api.npm === "@openrouter/ai-sdk-provider") return OpenRouter.configure(options).model(model.api.id)
+  throw new Error(`Native LLM request adapter does not support provider package ${model.api.npm}`)
+}
+
+export const request = (input: RequestInput) => {
+  const converted = messages(input.messages)
+  // This is the only native adapter boundary that should construct canonical
+  // @miaw/llm request objects from miaw's session/AI SDK-shaped data.
+  return LLM.request({
+    model: model(input, input.headers),
+    system: [...(input.system ?? []).map(SystemPart.make), ...converted.system],
+    messages: converted.messages,
+    tools: tools(input.tools),
+    toolChoice: input.toolChoice,
+    generation: generation(input),
+    providerOptions: input.providerOptions,
+  })
+}
+
+export * as LLMNative from "./native-request"

package/src/session/llm/native-runtime.ts

@@ -0,0 +1,195 @@
+import type { Auth } from "@/auth"
+import type { Provider } from "@/provider/provider"
+import { ProviderTransform } from "@/provider/transform"
+import { errorMessage } from "@/util/error"
+import { isRecord } from "@/util/record"
+import { asSchema, type ModelMessage, type Tool } from "ai"
+import { Cause, Effect, FiberSet, Queue } from "effect"
+import * as Stream from "effect/Stream"
+import { FetchHttpClient } from "effect/unstable/http"
+import {
+  LLMRequest,
+  Tool as NativeTool,
+  ToolFailure,
+  ToolRuntime,
+  toDefinitions,
+  type JsonSchema,
+  type LLMEvent,
+} from "@miaw/llm"
+import type { LLMClientShape } from "@miaw/llm/route"
+import { LLMNative } from "./native-request"
+
+export type RuntimeStatus =
+  | { readonly type: "supported"; readonly apiKey: string; readonly baseURL?: string }
+  | { readonly type: "unsupported"; readonly reason: string }
+export type StreamResult =
+  | { readonly type: "supported"; readonly stream: Stream.Stream<LLMEvent, unknown> }
+  | { readonly type: "unsupported"; readonly reason: string }
+
+type StreamInput = {
+  readonly model: Provider.Model
+  readonly provider: Provider.Info
+  readonly auth: Auth.Info | undefined
+  readonly llmClient: LLMClientShape
+  readonly messages: ModelMessage[]
+  readonly tools: Record<string, Tool>
+  readonly toolChoice?: "auto" | "required" | "none"
+  readonly temperature?: number
+  readonly topP?: number
+  readonly topK?: number
+  readonly maxOutputTokens?: number
+  readonly providerOptions?: Record<string, any>
+  readonly headers: Record<string, string>
+  readonly abort: AbortSignal
+}
+
+export function status(input: Pick<StreamInput, "model" | "provider" | "auth">): RuntimeStatus {
+  return statusWithFetch(input, providerFetch(input))
+}
+
+function statusWithFetch(
+  input: Pick<StreamInput, "model" | "provider" | "auth">,
+  fetch: typeof globalThis.fetch | undefined,
+): RuntimeStatus {
+  const providerID = input.model.providerID
+  if (providerID !== "openai" && providerID !== "anthropic" && !providerID.startsWith("miaw"))
+    return { type: "unsupported", reason: "provider is not openai, miaw, or anthropic" }
+  const npm = input.model.api.npm
+  if (npm !== "@ai-sdk/openai" && npm !== "@ai-sdk/openai-compatible" && npm !== "@ai-sdk/anthropic")
+    return { type: "unsupported", reason: "provider package is not OpenAI, OpenAI-compatible, or Anthropic" }
+  if (input.auth?.type === "oauth" && !(input.provider.id === "openai" && fetch)) {
+    return { type: "unsupported", reason: "OAuth auth requires a provider fetch override" }
+  }
+
+  const apiKey = typeof input.provider.options.apiKey === "string" ? input.provider.options.apiKey : input.provider.key
+  if (!apiKey) return { type: "unsupported", reason: "API key is not configured" }
+
+  return {
+    type: "supported",
+    apiKey,
+    baseURL: typeof input.provider.options.baseURL === "string" ? input.provider.options.baseURL : undefined,
+  }
+}
+
+export function stream(input: StreamInput): StreamResult {
+  const fetch = providerFetch(input)
+  const current = statusWithFetch(input, fetch)
+  if (current.type === "unsupported") return current
+
+  // Integration point with @miaw/llm: native-request lowers session data
+  // into an LLMRequest, then LLMClient handles route selection and transport.
+  //
+  // ProviderTransform.providerOptions builds AI-SDK-shaped options for the
+  // selected SDK key (e.g. "openai") and the native LLM SDK reads the same
+  // keys via OpenAIOptions.* (store, reasoningEffort, reasoningSummary,
+  // include, textVerbosity, promptCacheKey). Both sides intentionally use
+  // OpenAI's official wire field names, so this is identity, not translation
+  // — if a field ever needs to differ between the two surfaces, the
+  // translation belongs here, not split across both packages.
+  const tools = nativeTools(input.tools, input)
+  const request = LLMNative.request({
+    model: input.model,
+    apiKey: current.apiKey,
+    baseURL: current.baseURL,
+    messages: ProviderTransform.message(input.messages, input.model, input.providerOptions ?? {}),
+    toolChoice: input.toolChoice,
+    temperature: input.temperature,
+    topP: input.topP,
+    topK: input.topK,
+    maxOutputTokens: input.maxOutputTokens,
+    providerOptions: ProviderTransform.providerOptions(input.model, input.providerOptions ?? {}),
+    headers: { ...providerHeaders(input.provider.options.headers), ...input.headers },
+  })
+  const stream = Stream.scoped(
+    Stream.unwrap(
+      Effect.gen(function* () {
+        const settlements = yield* FiberSet.make<void>()
+        const results = yield* Queue.unbounded<LLMEvent, Cause.Done>()
+        const provider = input.llmClient
+          .stream(
+            LLMRequest.update(request, {
+              tools: [...request.tools, ...toDefinitions(tools)],
+            }),
+          )
+          .pipe(
+            Stream.flatMap((event) =>
+              event.type !== "tool-call" || event.providerExecuted
+                ? Stream.make(event)
+                : Stream.make(event).pipe(
+                    Stream.concat(
+                      Stream.fromEffectDrain(
+                        ToolRuntime.dispatch(tools, event).pipe(
+                          Effect.flatMap((dispatched) => Queue.offerAll(results, dispatched.events)),
+                          Effect.catchCause((cause) => Queue.failCause(results, cause)),
+                          Effect.asVoid,
+                          FiberSet.run(settlements, { startImmediately: true }),
+                        ),
+                      ),
+                    ),
+                  ),
+            ),
+            Stream.concat(
+              Stream.fromEffectDrain(
+                FiberSet.awaitEmpty(settlements).pipe(Effect.andThen(Queue.end(results)), Effect.asVoid),
+              ),
+            ),
+          )
+        return provider.pipe(Stream.concat(Stream.fromQueue(results)))
+      }),
+    ),
+  )
+
+  return {
+    ...current,
+    stream: fetch ? stream.pipe(Stream.provideService(FetchHttpClient.Fetch, fetch)) : stream,
+  }
+}
+
+function providerFetch(input: Pick<StreamInput, "provider" | "auth">): typeof globalThis.fetch | undefined {
+  if (input.provider.id !== "openai" || input.auth?.type !== "oauth") return undefined
+  const value: unknown = input.provider.options.fetch
+  if (typeof value !== "function") return undefined
+  return value as typeof globalThis.fetch
+}
+
+function providerHeaders(value: unknown): Record<string, string> | undefined {
+  if (!isRecord(value)) return undefined
+  return Object.fromEntries(
+    Object.entries(value).filter((entry): entry is [string, string] => typeof entry[1] === "string"),
+  )
+}
+
+function nativeSchema(value: unknown): JsonSchema {
+  if (!value || typeof value !== "object") return { type: "object", properties: {} }
+  if ("jsonSchema" in value && value.jsonSchema && typeof value.jsonSchema === "object")
+    return value.jsonSchema as JsonSchema
+  return asSchema(value as Parameters<typeof asSchema>[0]).jsonSchema as JsonSchema
+}
+
+export function nativeTools(tools: Record<string, Tool>, input: Pick<StreamInput, "messages" | "abort">) {
+  return Object.fromEntries(
+    Object.entries(tools).map(([name, item]) => [
+      name,
+      // Tool execution remains miaw-owned. The native runtime only adapts
+      // the @miaw/llm tool call back into the AI SDK Tool.execute shape.
+      NativeTool.make({
+        description: item.description ?? "",
+        jsonSchema: nativeSchema(item.inputSchema),
+        execute: (args: unknown, ctx) =>
+          Effect.tryPromise({
+            try: () => {
+              if (!item.execute) throw new Error(`Tool has no execute handler: ${name}`)
+              return item.execute(args, {
+                toolCallId: ctx?.id ?? name,
+                messages: input.messages,
+                abortSignal: input.abort,
+              })
+            },
+            catch: (error) => new ToolFailure({ message: errorMessage(error), error }),
+          }),
+      }),
+    ]),
+  )
+}
+
+export * as LLMNativeRuntime from "./native-runtime"

package/src/session/llm/request.ts

@@ -0,0 +1,216 @@
+import { PermissionV1 } from "@miaw/core/v1/permission"
+import type { Auth } from "@/auth"
+import { SessionV1 } from "@miaw/core/v1/session"
+import type { RuntimeFlags } from "@/effect/runtime-flags"
+import { InstanceState } from "@/effect/instance-state"
+import { Permission } from "@/permission"
+import type { Agent } from "@/agent/agent"
+import type { MessageV2 } from "../message-v2"
+import type { Provider } from "@/provider/provider"
+import { ProviderTransform } from "@/provider/transform"
+import { SystemPrompt } from "../system"
+import { InstallationVersion } from "@miaw/core/installation/version"
+import { Effect, Record } from "effect"
+import { jsonSchema, tool as aiTool, type ModelMessage, type Tool } from "ai"
+import type { Plugin } from "@/plugin"
+import { mergeDeep } from "remeda"
+
+const USER_AGENT = `miaw/${InstallationVersion}`
+
+type PrepareInput = {
+  readonly user: SessionV1.User
+  readonly sessionID: string
+  readonly parentSessionID?: string
+  readonly model: Provider.Model
+  readonly agent: Agent.Info
+  readonly permission?: PermissionV1.Ruleset
+  readonly system: string[]
+  readonly messages: ModelMessage[]
+  readonly small?: boolean
+  readonly tools: Record<string, Tool>
+  readonly provider: Provider.Info
+  readonly auth: Auth.Info | undefined
+  readonly plugin: Plugin.Interface
+  readonly flags: RuntimeFlags.Info
+  readonly isWorkflow: boolean
+}
+
+export type Prepared = {
+  readonly system: string[]
+  readonly messages: ModelMessage[]
+  readonly tools: Record<string, Tool>
+  readonly params: {
+    readonly temperature?: number
+    readonly topP?: number
+    readonly topK?: number
+    readonly maxOutputTokens?: number
+    readonly options: Record<string, any>
+  }
+  readonly messageTransformOptions: Record<string, any>
+  readonly headers: Record<string, string>
+}
+
+const mergeOptions = (target: Record<string, any>, source: Record<string, any> | undefined): Record<string, any> =>
+  mergeDeep(target, source ?? {}) as Record<string, any>
+
+export const prepare = Effect.fn("LLMRequestPrep.prepare")(function* (input: PrepareInput) {
+  const isOpenaiOauth = input.provider.id === "openai" && input.auth?.type === "oauth"
+  const system = [
+    [
+      ...(input.agent.prompt ? [input.agent.prompt] : SystemPrompt.provider(input.model)),
+      ...input.system,
+      ...(input.user.system ? [input.user.system] : []),
+    ]
+      .filter((x) => x)
+      .join("\n"),
+  ]
+
+  const header = system[0]
+  yield* input.plugin.trigger(
+    "experimental.chat.system.transform",
+    { sessionID: input.sessionID, model: input.model },
+    { system },
+  )
+  if (system.length > 2 && system[0] === header) {
+    const rest = system.slice(1)
+    system.length = 0
+    system.push(header, rest.join("\n"))
+  }
+
+  const variant =
+    !input.small && input.model.variants && input.user.model.variant
+      ? input.model.variants[input.user.model.variant]
+      : {}
+  const base = input.small
+    ? ProviderTransform.smallOptions(input.model)
+    : ProviderTransform.options({
+        model: input.model,
+        sessionID: input.sessionID,
+        providerOptions: input.provider.options,
+      })
+  const options = mergeOptions(mergeOptions(mergeOptions(base, input.model.options), input.agent.options), variant)
+  if (
+    input.model.api.npm === "@ai-sdk/azure" &&
+    (input.provider.options.useCompletionUrls || input.model.options.useCompletionUrls || options.useCompletionUrls)
+  ) {
+    delete options.reasoningSummary
+    delete options.include
+  }
+  if (isOpenaiOauth) options.instructions = system.join("\n")
+
+  const messages =
+    isOpenaiOauth || input.isWorkflow
+      ? input.messages
+      : [
+          ...system.map(
+            (x): ModelMessage => ({
+              role: "system",
+              content: x,
+            }),
+          ),
+          ...input.messages,
+        ]
+
+  const params = yield* input.plugin.trigger(
+    "chat.params",
+    {
+      sessionID: input.sessionID,
+      agent: input.agent.name,
+      model: input.model,
+      provider: input.provider,
+      message: input.user,
+    },
+    {
+      temperature: input.model.capabilities.temperature
+        ? (input.agent.temperature ?? ProviderTransform.temperature(input.model))
+        : undefined,
+      topP: input.agent.topP ?? ProviderTransform.topP(input.model),
+      topK: ProviderTransform.topK(input.model),
+      maxOutputTokens: ProviderTransform.maxOutputTokens(input.model, input.flags.outputTokenMax),
+      options,
+    },
+  )
+
+  const { headers } = yield* input.plugin.trigger(
+    "chat.headers",
+    {
+      sessionID: input.sessionID,
+      agent: input.agent.name,
+      model: input.model,
+      provider: input.provider,
+      message: input.user,
+    },
+    {
+      headers: {},
+    },
+  )
+
+  const tools = resolveTools(input)
+  if (
+    input.model.providerID.includes("github-copilot") &&
+    Object.keys(tools).length === 0 &&
+    hasToolCalls(input.messages)
+  ) {
+    // Copilot needs a tools field when replaying prior tool calls, even if no tools are currently enabled.
+    tools["_noop"] = aiTool({
+      description: "Do not call this tool. It exists only for API compatibility and must never be invoked.",
+      inputSchema: jsonSchema({
+        type: "object",
+        properties: {
+          reason: { type: "string", description: "Unused" },
+        },
+      }),
+      execute: async () => ({ output: "", title: "", metadata: {} }),
+    })
+  }
+
+  const miawProjectID = input.model.providerID.startsWith("miaw")
+    ? (yield* InstanceState.context).project.id
+    : undefined
+
+  return {
+    system,
+    messages,
+    tools: Object.fromEntries(Object.entries(tools).toSorted(([a], [b]) => a.localeCompare(b))),
+    params,
+    messageTransformOptions: options,
+    headers: {
+      ...(input.model.providerID.startsWith("miaw")
+        ? {
+            ...(miawProjectID ? { "x-miaw-project": miawProjectID } : {}),
+            "x-miaw-session": input.sessionID,
+            "x-miaw-request": input.user.id,
+            "x-miaw-client": input.flags.client,
+            "User-Agent": USER_AGENT,
+          }
+        : {
+            "x-session-affinity": input.sessionID,
+            "X-Session-Id": input.sessionID,
+            ...(input.parentSessionID ? { "x-parent-session-id": input.parentSessionID } : {}),
+            "User-Agent": USER_AGENT,
+          }),
+      ...input.model.headers,
+      ...headers,
+    },
+  }
+})
+
+function resolveTools(input: Pick<PrepareInput, "tools" | "agent" | "permission" | "user">) {
+  const disabled = Permission.disabled(
+    Object.keys(input.tools),
+    Permission.merge(input.agent.permission, input.permission ?? []),
+  )
+  return Record.filter(input.tools, (_, k) => input.user.tools?.[k] !== false && !disabled.has(k))
+}
+
+export function hasToolCalls(messages: ModelMessage[]): boolean {
+  for (const msg of messages) {
+    if (!Array.isArray(msg.content)) continue
+    for (const part of msg.content) {
+      if (part.type === "tool-call" || part.type === "tool-result") return true
+    }
+  }
+  return false
+}
+
+export * as LLMRequestPrep from "./request"