feat(opencode): cache-friendly compaction via primary loop request path

Default compaction no longer uses a dedicated compaction agent. It inherits the session's active agent and delegates request assembly to the primary loop's shared envelope (same system prompt, instructions, skills, MCP context, and tool definitions), replaying the typed message head and appending the summary instruction as the final user message. This keeps the compaction request prefix-identical to the preceding turn so provider prompt/KV caches are reused instead of fully re-prefilled. An explicitly configured agent.compaction preserves the legacy behavior exactly: dedicated hidden agent, optional model override, and the serialized-transcript summary request.
2026-08-14 23:38:23 -04:00 · 2026-08-13 23:10:07 -05:00
29 changed files with 299 additions and 337 deletions
@@ -173,6 +173,18 @@ export const buildPrompt = (input: { readonly previousSummary?: string; readonly
  ].join("\n\n")
 }

+export const buildReplayPrompt = (previousSummary?: string) => {
+  const instruction =
+    "Create a new anchored summary from the conversation messages above so another coding agent can continue the work."
+  if (!previousSummary) return [instruction, SUMMARY_TEMPLATE].join("\n\n")
+  return [
+    instruction,
+    `Here is the summary of the conversation before the messages above:\n\n<prior-summary>\n${previousSummary}\n</prior-summary>`,
+    SUMMARY_UPDATE_INSTRUCTIONS,
+    SUMMARY_TEMPLATE,
+  ].join("\n\n")
+}
+
 export const make = (dependencies: Dependencies) => {
  const config = settings(dependencies.config)
  const compactAfterOverflow = Effect.fn("SessionCompaction.compactAfterOverflow")(function* (input: Input) {
@@ -216,21 +216,6 @@ const layer = Layer.effect(
            mode: "subagent",
            native: true,
          },
-          compaction: {
-            name: "compaction",
-            mode: "primary",
-            native: true,
-            hidden: true,
-            prompt: PROMPT_COMPACTION,
-            permission: Permission.merge(
-              defaults,
-              Permission.fromConfig({
-                "*": "deny",
-              }),
-              user,
-            ),
-            options: {},
-          },
          title: {
            name: "title",
            mode: "primary",
@@ -270,6 +255,26 @@ const layer = Layer.effect(
            continue
          }
          let item = agents[key]
+          // The compaction agent is not registered by default: compaction inherits
+          // the session's active agent so its request stays prefix-identical for
+          // prompt caching. An explicit `agent.compaction` config opts into the
+          // legacy dedicated agent, seeded here so overrides merge as before.
+          if (!item && key === "compaction")
+            item = agents[key] = {
+              name: "compaction",
+              mode: "primary",
+              native: true,
+              hidden: true,
+              prompt: PROMPT_COMPACTION,
+              permission: Permission.merge(
+                defaults,
+                Permission.fromConfig({
+                  "*": "deny",
+                }),
+                user,
+              ),
+              options: {},
+            }
          if (!item)
            item = agents[key] = {
              name: key,
@@ -37,12 +37,10 @@ function base64UrlEncode(buffer: ArrayBuffer): string {

 export interface IdTokenClaims {
  chatgpt_account_id?: string
-  chatgpt_compute_residency?: string
  organizations?: Array<{ id: string }>
  email?: string
  "https://api.openai.com/auth"?: {
    chatgpt_account_id?: string
-    chatgpt_compute_residency?: string
  }
 }

@@ -77,14 +75,6 @@ export function extractAccountId(tokens: TokenResponse): string | undefined {
  return undefined
 }

-export function extractResidency(token: string): string | undefined {
-  const claims = parseJwtClaims(token)
-  const residency =
-    claims?.["https://api.openai.com/auth"]?.chatgpt_compute_residency ?? claims?.chatgpt_compute_residency
-  if (!residency || residency === "no_constraint") return undefined
-  return residency
-}
-
 function buildAuthorizeUrl(redirectUri: string, pkce: PkceCodes, state: string): string {
  const params = new URLSearchParams({
    response_type: "code",
@@ -421,12 +411,10 @@ export async function CodexAuthPlugin(input: PluginInput, options: CodexAuthPlug
              requestInput instanceof URL
                ? requestInput
                : new URL(typeof requestInput === "string" ? requestInput : requestInput.url)
-            const rewrite = parsed.pathname.includes("/v1/responses") || parsed.pathname.includes("/chat/completions")
-            const url = rewrite ? new URL(codexApiEndpoint) : parsed
-            if (rewrite) {
-              const residency = extractResidency(currentAuth.access)
-              if (residency) headers.set("x-openai-internal-codex-residency", residency)
-            }
+            const url =
+              parsed.pathname.includes("/v1/responses") || parsed.pathname.includes("/chat/completions")
+                ? new URL(codexApiEndpoint)
+                : parsed

            const requestInit = {
              ...init,
@@ -20,7 +20,7 @@ import { RuntimeFlags } from "@/effect/runtime-flags"
 import { EventV2Bridge } from "@/event-v2-bridge"
 import { ProviderV2 } from "@opencode-ai/core/provider"
 import { ModelV2 } from "@opencode-ai/core/model"
-import { buildPrompt } from "@opencode-ai/core/session/compaction"
+import { buildPrompt, buildReplayPrompt } from "@opencode-ai/core/session/compaction"
 import { SessionCompactionEvent } from "@opencode-ai/schema/session-compaction-event"

 export const Event = SessionCompactionEvent
@@ -162,6 +162,15 @@ function splitTurn(input: {
  })
 }

+export type Run = (input: {
+  user: SessionV1.User
+  agent: Agent.Info
+  model: Provider.Model
+  processor: SessionProcessor.Handle
+  messages: SessionV1.WithParts[]
+  prompt: string
+}) => Effect.Effect<SessionProcessor.Result>
+
 export interface Interface {
  readonly isOverflow: (input: {
    tokens: SessionV1.Assistant["tokens"]
@@ -174,6 +183,7 @@ export interface Interface {
    sessionID: SessionID
    auto: boolean
    overflow?: boolean
+    run?: Run
  }) => Effect.Effect<"continue" | "stop">
  readonly create: (input: {
    sessionID: SessionID
@@ -322,6 +332,7 @@ const layer = Layer.effect(
      sessionID: SessionID
      auto: boolean
      overflow?: boolean
+      run?: Run
    }) {
      const parent = input.messages.findLast((m) => m.info.id === input.parentID)
      if (!parent || parent.info.role !== "user") {
@@ -355,11 +366,13 @@ const layer = Layer.effect(
        }
      }

-      const agent = yield* agents.get("compaction")
-      const model = agent.model
-        ? yield* provider.getModel(agent.model.providerID, agent.model.modelID).pipe(Effect.orDie)
-        : yield* provider.getModel(userMessage.model.providerID, userMessage.model.modelID).pipe(Effect.orDie)
      const cfg = yield* config.get()
+      const configured = cfg.agent?.compaction !== undefined
+      const agent = yield* agents.get(configured ? "compaction" : userMessage.agent)
+      const model =
+        configured && agent.model
+          ? yield* provider.getModel(agent.model.providerID, agent.model.modelID).pipe(Effect.orDie)
+          : yield* provider.getModel(userMessage.model.providerID, userMessage.model.modelID).pipe(Effect.orDie)
      const history = compactionPart && messages.at(-1)?.info.id === input.parentID ? messages.slice(0, -1) : messages
      const prior = completedCompactions(history)
      const hidden = new Set(prior.flatMap((item) => [item.userIndex, item.assistantIndex]))
@@ -377,18 +390,16 @@ const layer = Layer.effect(
      )
      const msgs = structuredClone(selected.head)
      yield* plugin.trigger("experimental.chat.messages.transform", {}, { messages: msgs })
-      const conversation = msgs.map(serialize).filter(Boolean).join("\n\n")
-      const nextPrompt =
+      // A configured compaction agent keeps the legacy serialized-transcript path.
+      const run = configured ? undefined : input.run
+      const conversation = run ? "" : msgs.map(serialize).filter(Boolean).join("\n\n")
+      const nextPrompt = [
        compacting.prompt ??
-        [
-          buildPrompt({
-            previousSummary,
-            context: [conversation],
-          }),
-          ...compacting.context,
-        ]
-          .filter(Boolean)
-          .join("\n\n")
+          (run ? buildReplayPrompt(previousSummary) : buildPrompt({ previousSummary, context: [conversation] })),
+        ...compacting.context,
+      ]
+        .filter(Boolean)
+        .join("\n\n")
      const ctx = yield* InstanceState.context
      const msg: SessionV1.Assistant = {
        id: MessageID.ascending(),
@@ -422,30 +433,43 @@ const layer = Layer.effect(
        sessionID: input.sessionID,
        model,
      })
-      const result = yield* processor.process({
-        user: userMessage,
-        agent,
-        sessionID: input.sessionID,
-        tools: {},
-        system: [],
-        messages: [
-          {
-            role: "user",
-            content: [
+      // The synthetic compaction marker carries no per-turn system/tools settings.
+      const active = msgs.findLast(
+        (message) => message.info.role === "user" && !message.parts.some((part) => part.type === "compaction"),
+      )
+      const result = run
+        ? yield* run({
+            user: active?.info.role === "user" ? active.info : userMessage,
+            agent,
+            model,
+            processor,
+            messages: msgs,
+            prompt: nextPrompt,
+          })
+        : yield* processor.process({
+            user: userMessage,
+            agent,
+            sessionID: input.sessionID,
+            tools: {},
+            system: [],
+            messages: [
              {
-                type: "text",
-                text: [
-                  nextPrompt,
-                  ...(compacting.prompt ? ["The following is the conversation history:", conversation] : []),
-                ]
-                  .filter(Boolean)
-                  .join("\n\n"),
+                role: "user",
+                content: [
+                  {
+                    type: "text",
+                    text: [
+                      nextPrompt,
+                      ...(compacting.prompt ? ["The following is the conversation history:", conversation] : []),
+                    ]
+                      .filter(Boolean)
+                      .join("\n\n"),
+                  },
+                ],
              },
            ],
-          },
-        ],
-        model,
-      })
+            model,
+          })

      if (result === "compact") {
        processor.message.error = new SessionV1.ContextOverflowError({
@@ -149,6 +149,45 @@ const layer = Layer.effect(
      } satisfies TaskPromptOps
    })

+    // One assembly path keeps compaction requests prefix-identical to normal turns for prompt caching.
+    const prepare = Effect.fn("SessionPrompt.prepare")(function* (input: {
+      session: Session.Info
+      agent: Agent.Info
+      model: Provider.Model
+      processor: SessionProcessor.Handle
+      messages: SessionV1.WithParts[]
+    }) {
+      const lastUserMsg = input.messages.findLast((message) => message.info.role === "user")
+      const tools = yield* SessionTools.resolve({
+        agent: input.agent,
+        session: input.session,
+        model: input.model,
+        processor: input.processor,
+        bypassAgentCheck: lastUserMsg?.parts.some((part) => part.type === "agent") ?? false,
+        messages: input.messages,
+        promptOps: yield* ops(),
+      }).pipe(
+        Effect.provideService(Plugin.Service, plugin),
+        Effect.provideService(Permission.Service, permission),
+        Effect.provideService(ToolRegistry.Service, registry),
+        Effect.provideService(MCP.Service, mcp),
+        Effect.provideService(Truncate.Service, truncate),
+        Effect.provideService(RuntimeFlags.Service, flags),
+      )
+      const [skills, env, instructions, mcpInstructions, modelMessages] = yield* Effect.all([
+        sys.skills(input.agent),
+        sys.environment(input.model),
+        instruction.system().pipe(Effect.orDie),
+        sys.mcp(input.agent, input.session.permission),
+        MessageV2.toModelMessagesEffect(input.messages, input.model),
+      ])
+      return {
+        tools,
+        modelMessages,
+        system: [...env, ...instructions, ...(mcpInstructions ? [mcpInstructions] : []), ...(skills ? [skills] : [])],
+      }
+    })
+
    const cancel = Effect.fn("SessionPrompt.cancel")(function* (sessionID: SessionID) {
      yield* Effect.logInfo("cancel", { "session.id": sessionID })
      yield* state.cancel(sessionID)
@@ -1153,6 +1192,21 @@ const layer = Layer.effect(
              sessionID,
              auto: task.auto,
              overflow: task.overflow,
+              run: ({ user, agent, model, processor, messages, prompt }) =>
+                Effect.gen(function* () {
+                  const prepared = yield* prepare({ session, agent, model, processor, messages })
+                  return yield* processor.process({
+                    user,
+                    agent,
+                    permission: session.permission,
+                    sessionID,
+                    parentSessionID: session.parentID,
+                    system: prepared.system,
+                    messages: [...prepared.modelMessages, { role: "user", content: prompt }],
+                    tools: prepared.tools,
+                    model,
+                  })
+                }),
            })
            if (result === "stop") break
            continue
@@ -1219,68 +1273,34 @@ const layer = Layer.effect(
            .pipe(Effect.onInterrupt(() => finalizeInterruptedAssistant))

          const outcome: "break" | "continue" = yield* Effect.gen(function* () {
-            const lastUserMsg = msgs.findLast((m) => m.info.role === "user")
-            const bypassAgentCheck = lastUserMsg?.parts.some((p) => p.type === "agent") ?? false
-            const promptOps = yield* ops()
-
-            const tools = yield* SessionTools.resolve({
-              agent,
-              session,
-              model,
-              processor: handle,
-              bypassAgentCheck,
-              messages: msgs,
-              promptOps,
-            }).pipe(
-              Effect.provideService(Plugin.Service, plugin),
-              Effect.provideService(Permission.Service, permission),
-              Effect.provideService(ToolRegistry.Service, registry),
-              Effect.provideService(MCP.Service, mcp),
-              Effect.provideService(Truncate.Service, truncate),
-              Effect.provideService(RuntimeFlags.Service, flags),
-            )
-
-            if (lastUser.format?.type === "json_schema") {
-              tools["StructuredOutput"] = createStructuredOutputTool({
-                schema: lastUser.format.schema,
-                onSuccess(output) {
-                  structured = output
-                },
-              })
-            }
-
            if (step === 1)
              yield* summary.summarize({ sessionID, messageID: lastUser.id }).pipe(Effect.ignore, Effect.forkIn(scope))

            yield* plugin.trigger("experimental.chat.messages.transform", {}, { messages: msgs })

-            const [skills, env, instructions, mcpInstructions, modelMsgs] = yield* Effect.all([
-              sys.skills(agent),
-              sys.environment(model),
-              instruction.system().pipe(Effect.orDie),
-              sys.mcp(agent, session.permission),
-              MessageV2.toModelMessagesEffect(msgs, model),
-            ])
-            const system = [
-              ...env,
-              ...instructions,
-              ...(mcpInstructions ? [mcpInstructions] : []),
-              ...(skills ? [skills] : []),
-            ]
+            const prepared = yield* prepare({ session, agent, model, processor: handle, messages: msgs })
            const format = lastUser.format ?? { type: "text" as const }
-            if (format.type === "json_schema") system.push(STRUCTURED_OUTPUT_SYSTEM_PROMPT)
+            if (format.type === "json_schema") {
+              prepared.system.push(STRUCTURED_OUTPUT_SYSTEM_PROMPT)
+              prepared.tools["StructuredOutput"] = createStructuredOutputTool({
+                schema: format.schema,
+                onSuccess(output) {
+                  structured = output
+                },
+              })
+            }
            const result = yield* handle.process({
              user: lastUser,
              agent,
              permission: session.permission,
              sessionID,
              parentSessionID: session.parentID,
-              system,
+              system: prepared.system,
              messages: [
-                ...modelMsgs,
+                ...prepared.modelMessages,
                ...(isLastStep ? [{ role: "assistant" as const, content: MAX_STEPS_PROMPT }] : []),
              ],
-              tools,
+              tools: prepared.tools,
              model,
              toolChoice: format.type === "json_schema" ? "required" : undefined,
            })
@@ -52,7 +52,7 @@ it.instance("returns default native agents when no config", () =>
    expect(names).toContain("plan")
    expect(names).toContain("general")
    expect(names).toContain("explore")
-    expect(names).toContain("compaction")
+    expect(names).not.toContain("compaction")
    expect(names).toContain("title")
    expect(names).toContain("summary")
  }),
@@ -170,17 +170,34 @@ it.instance("general agent denies todo tools", () =>
  }),
 )

-it.instance("compaction agent denies all permissions", () =>
+it.instance("compaction agent is absent without configuration", () =>
  Effect.gen(function* () {
    const compaction = yield* load((svc) => svc.get("compaction"))
-    expect(compaction).toBeDefined()
-    expect(compaction?.hidden).toBe(true)
-    expect(evalPerm(compaction, "bash")).toBe("deny")
-    expect(evalPerm(compaction, "edit")).toBe("deny")
-    expect(evalPerm(compaction, "read")).toBe("deny")
+    expect(compaction).toBeUndefined()
  }),
 )

+it.instance(
+  "configured compaction agent keeps native defaults",
+  () =>
+    Effect.gen(function* () {
+      const compaction = yield* load((svc) => svc.get("compaction"))
+      expect(compaction).toBeDefined()
+      expect(compaction?.native).toBe(true)
+      expect(compaction?.hidden).toBe(true)
+      expect(evalPerm(compaction, "bash")).toBe("deny")
+      expect(evalPerm(compaction, "edit")).toBe("deny")
+      expect(evalPerm(compaction, "read")).toBe("deny")
+    }),
+  {
+    config: {
+      agent: {
+        compaction: {},
+      },
+    },
+  },
+)
+
 it.instance(
  "custom agent from config creates new agent",
  () =>
@@ -710,6 +727,9 @@ it.instance(
  {
    config: {
      default_agent: "compaction",
+      agent: {
+        compaction: {},
+      },
    },
  },
 )
@@ -1,13 +1,9 @@
 import { describe, expect, test } from "bun:test"
-import { createServer, type IncomingMessage } from "node:http"
-import { type AddressInfo } from "node:net"
-import { WebSocketServer } from "ws"
 import {
  CodexAuthPlugin,
  parseJwtClaims,
  extractAccountIdFromClaims,
  extractAccountId,
-  extractResidency,
  renderOAuthError,
  type IdTokenClaims,
 } from "../../src/plugin/openai/codex"
@@ -135,69 +131,6 @@ describe("plugin.codex", () => {
    })
  })

-  describe("extractResidency", () => {
-    test("extracts compute residency from the namespaced auth claims", () => {
-      expect(
-        extractResidency(
-          createTestJwt({
-            "https://api.openai.com/auth": { chatgpt_compute_residency: "eu" },
-          }),
-        ),
-      ).toBe("eu")
-    })
-
-    test("falls back to a root compute residency claim", () => {
-      expect(extractResidency(createTestJwt({ chatgpt_compute_residency: "us" }))).toBe("us")
-    })
-
-    test("supports compute residency values without maintaining a region list", () => {
-      expect(
-        extractResidency(
-          createTestJwt({
-            "https://api.openai.com/auth": { chatgpt_compute_residency: "ae" },
-          }),
-        ),
-      ).toBe("ae")
-      expect(
-        extractResidency(
-          createTestJwt({
-            "https://api.openai.com/auth": { chatgpt_compute_residency: "future-region_1" },
-          }),
-        ),
-      ).toBe("future-region_1")
-    })
-
-    test("ignores unconstrained and data residency values", () => {
-      expect(
-        extractResidency(
-          createTestJwt({
-            "https://api.openai.com/auth": { chatgpt_compute_residency: "no_constraint" },
-          }),
-        ),
-      ).toBeUndefined()
-      expect(
-        extractResidency(
-          createTestJwt({
-            "https://api.openai.com/auth": { chatgpt_data_residency: "gb" },
-          }),
-        ),
-      ).toBeUndefined()
-      expect(extractResidency(createTestJwt({ chatgpt_compute_residency: "" }))).toBeUndefined()
-      expect(extractResidency("not-a-jwt")).toBeUndefined()
-    })
-
-    test("prefers a namespaced unconstrained value over a root residency", () => {
-      expect(
-        extractResidency(
-          createTestJwt({
-            chatgpt_compute_residency: "eu",
-            "https://api.openai.com/auth": { chatgpt_compute_residency: "no_constraint" },
-          }),
-        ),
-      ).toBeUndefined()
-    })
-  })
-
  test("installs websocket transport only when experimental websockets are enabled", async () => {
    const disabled = await CodexAuthPlugin({} as never)
    const enabled = await CodexAuthPlugin({} as never, { experimentalWebSockets: true })
@@ -216,73 +149,6 @@ describe("plugin.codex", () => {
    await enabled.dispose?.()
  })

-  test("sends token residency only to the ChatGPT Codex backend", async () => {
-    const requests: Array<{ path: string; residency: string | null }> = []
-    using server = Bun.serve({
-      port: 0,
-      fetch(request) {
-        requests.push({
-          path: new URL(request.url).pathname,
-          residency: request.headers.get("x-openai-internal-codex-residency"),
-        })
-        return new Response("{}")
-      },
-    })
-    const hooks = await CodexAuthPlugin({} as never, {
-      codexApiEndpoint: new URL("/backend-api/codex/responses", server.url).toString(),
-    })
-    const loaded = await hooks.auth!.loader!(
-      async () =>
-        ({
-          type: "oauth",
-          refresh: "refresh",
-          access: createTestJwt({
-            "https://api.openai.com/auth": { chatgpt_compute_residency: "eu" },
-          }),
-          expires: Date.now() + 60_000,
-        }) as never,
-      {} as never,
-    )
-
-    await loaded.fetch!("https://api.openai.com/v1/responses")
-    await loaded.fetch!(new URL("/other", server.url))
-
-    expect(requests).toEqual([
-      { path: "/backend-api/codex/responses", residency: "eu" },
-      { path: "/other", residency: null },
-    ])
-  })
-
-  test("sends token residency through the WebSocket transport", async () => {
-    await using server = await createCodexWebSocketServer()
-    const hooks = await CodexAuthPlugin({} as never, {
-      codexApiEndpoint: server.url,
-      experimentalWebSockets: true,
-    })
-    const loaded = await hooks.auth!.loader!(
-      async () =>
-        ({
-          type: "oauth",
-          refresh: "refresh",
-          access: createTestJwt({
-            "https://api.openai.com/auth": { chatgpt_compute_residency: "eu" },
-          }),
-          expires: Date.now() + 60_000,
-        }) as never,
-      {} as never,
-    )
-
-    const response = await loaded.fetch!("https://api.openai.com/v1/responses", {
-      method: "POST",
-      headers: { "session-id": "session-1" },
-      body: JSON.stringify({ stream: true, input: "hi" }),
-    })
-
-    expect(await response.text()).toContain("data: [DONE]")
-    expect(server.headers()?.["x-openai-internal-codex-residency"]).toBe("eu")
-    await hooks.dispose?.()
-  })
-
  test("filters unsupported modes and uses Codex context limits for OAuth GPT models", async () => {
    const hooks = await CodexAuthPlugin({} as never)
    const limit = { context: 1_050_000, input: 922_000, output: 128_000 }
@@ -327,9 +193,6 @@ describe("plugin.codex", () => {
  })

  test("deduplicates concurrent Codex token refreshes", async () => {
-    const refreshedAccess = createTestJwt({
-      "https://api.openai.com/auth": { chatgpt_compute_residency: "eu" },
-    })
    let auth = {
      type: "oauth" as const,
      refresh: "refresh-old",
@@ -344,7 +207,7 @@ describe("plugin.codex", () => {
      resolveRefresh = resolve
    })
    let refreshRequests = 0
-    const apiRequests: { authorization: string | null; accountId: string | null; residency: string | null }[] = []
+    const apiRequests: { authorization: string | null; accountId: string | null }[] = []

    using server = Bun.serve({
      port: 0,
@@ -356,7 +219,7 @@ describe("plugin.codex", () => {
          await refreshReady
          return Response.json({
            id_token: createTestJwt({ chatgpt_account_id: "acc-123" }),
-            access_token: refreshedAccess,
+            access_token: "access-new",
            refresh_token: "refresh-new",
            expires_in: 3600,
          })
@@ -366,7 +229,6 @@ describe("plugin.codex", () => {
          apiRequests.push({
            authorization: request.headers.get("authorization"),
            accountId: request.headers.get("ChatGPT-Account-Id"),
-            residency: request.headers.get("x-openai-internal-codex-residency"),
          })
          return new Response("{}", { status: 200 })
        }
@@ -419,11 +281,11 @@ describe("plugin.codex", () => {
    expect(refreshRequests).toBe(1)
    expect(authUpdates).toHaveLength(1)
    expect(authUpdates[0]?.body.refresh).toBe("refresh-new")
-    expect(authUpdates[0]?.body.access).toBe(refreshedAccess)
+    expect(authUpdates[0]?.body.access).toBe("access-new")
    expect(authUpdates[0]?.body.accountId).toBe("acc-123")
    expect(apiRequests).toEqual([
-      { authorization: `Bearer ${refreshedAccess}`, accountId: "acc-123", residency: "eu" },
-      { authorization: `Bearer ${refreshedAccess}`, accountId: "acc-123", residency: "eu" },
+      { authorization: "Bearer access-new", accountId: "acc-123" },
+      { authorization: "Bearer access-new", accountId: "acc-123" },
    ])
  })
 })
@@ -435,29 +297,3 @@ async function waitFor(predicate: () => boolean) {
    await new Promise((resolve) => setTimeout(resolve, 1))
  }
 }
-
-async function createCodexWebSocketServer() {
-  let headers: IncomingMessage["headers"] | undefined
-  const server = createServer()
-  const sockets = new WebSocketServer({ server })
-  sockets.on("connection", (socket, request) => {
-    headers = request.headers
-    socket.once("message", () => {
-      socket.send(JSON.stringify({ type: "response.completed", response: { id: "resp_123" } }))
-    })
-  })
-  await new Promise<void>((resolve, reject) => {
-    server.once("error", reject)
-    server.listen(0, "127.0.0.1", resolve)
-  })
-  const address = server.address() as AddressInfo
-  return {
-    url: `http://127.0.0.1:${address.port}/backend-api/codex/responses`,
-    headers: () => headers,
-    async [Symbol.asyncDispose]() {
-      for (const socket of sockets.clients) socket.terminate()
-      sockets.close()
-      server.close()
-    },
-  }
-}
@@ -17,18 +17,13 @@ describe("plugin.openai.ws", () => {

    const socket = await OpenAIWebSocket.connectResponsesWebSocket({
      url: server.wsUrl,
-      headers: {
-        authorization: "Bearer test",
-        "content-length": "123",
-        "x-openai-internal-codex-residency": "eu",
-      },
+      headers: { authorization: "Bearer test", "content-length": "123" },
    })

    expect(OpenAIWebSocket.toWebSocketUrl("http://example.com/v1/responses")).toBe("ws://example.com/v1/responses")
    expect(OpenAIWebSocket.toWebSocketUrl("https://example.com/v1/responses")).toBe("wss://example.com/v1/responses")
    expect(headers?.authorization).toBe("Bearer test")
    expect(headers?.["openai-beta"]).toBe(OpenAIWebSocket.PROTOCOL_HEADER)
-    expect(headers?.["x-openai-internal-codex-residency"]).toBe("eu")
    expect(headers?.["content-length"]).toBeUndefined()
    socket.terminate()
  })
@@ -222,6 +222,11 @@ function cfg(compaction?: ConfigV1.Info["compaction"]) {
  return Layer.succeed(Config.Service, TestConfig.make({ get: () => Effect.succeed({ ...base, compaction }) }))
 }

+function cfgAgent(agent: NonNullable<ConfigV1.Info["agent"]>) {
+  const config = Schema.decodeUnknownSync(ConfigV1.Info)({ agent }) as ConfigV1.Info
+  return Layer.succeed(Config.Service, TestConfig.make({ get: () => Effect.succeed(config) }))
+}
+
 const defaultProvider = wide()
 const compactionTestNode = LayerNode.group([
  SessionCompaction.node,
@@ -812,6 +817,75 @@ describe("session.compaction.prune", () => {
 })

 describe("session.compaction.process", () => {
+  itCompaction.instance(
+    "inherits the last agent when compaction is not configured",
+    () => {
+      const stub = llm()
+      let captured: LLM.StreamInput | undefined
+      stub.push(reply("summary", (input) => (captured = input)))
+      return Effect.gen(function* () {
+        const ssn = yield* SessionNs.Service
+        const session = yield* ssn.create({})
+        const msg = yield* createUserMessage(session.id, "hello")
+        const msgs = yield* ssn.messages({ sessionID: session.id })
+
+        yield* SessionCompaction.use.process({
+          parentID: msg.id,
+          messages: msgs,
+          sessionID: session.id,
+          auto: false,
+        })
+
+        expect(captured?.agent.name).toBe("build")
+        expect(captured?.agent.prompt).toBe("custom build prompt")
+        expect(captured?.agent.temperature).toBe(0.37)
+        expect(String(captured?.model.providerID)).toBe("test")
+        expect(String(captured?.model.id)).toBe("test-model")
+      }).pipe(
+        withCompaction({
+          llm: stub.llmLayer,
+          config: cfgAgent({
+            build: { prompt: "custom build prompt", temperature: 0.37, model: "missing/missing" },
+          }),
+        }),
+      )
+    },
+    { git: true },
+  )
+
+  itCompaction.instance(
+    "preserves an explicitly configured compaction agent",
+    () => {
+      const stub = llm()
+      let captured: LLM.StreamInput | undefined
+      stub.push(reply("summary", (input) => (captured = input)))
+      return Effect.gen(function* () {
+        const ssn = yield* SessionNs.Service
+        const session = yield* ssn.create({})
+        const msg = yield* createUserMessage(session.id, "hello")
+        const msgs = yield* ssn.messages({ sessionID: session.id })
+
+        yield* SessionCompaction.use.process({
+          parentID: msg.id,
+          messages: msgs,
+          sessionID: session.id,
+          auto: false,
+        })
+
+        expect(captured?.agent.name).toBe("compaction")
+        expect(captured?.agent.prompt).toBe("custom compaction prompt")
+        expect(captured?.agent.temperature).toBe(0.81)
+        expect(captured?.agent.hidden).toBe(true)
+      }).pipe(
+        withCompaction({
+          llm: stub.llmLayer,
+          config: cfgAgent({ compaction: { prompt: "custom compaction prompt", temperature: 0.81 } }),
+        }),
+      )
+    },
+    { git: true },
+  )
+
  it.instance(
    "throws when parent is not a user message",
    Effect.gen(function* () {
@@ -444,6 +444,34 @@ const boot = Effect.fn("test.boot")(function* (input?: { title?: string }) {

 // Loop semantics

+it.instance("default compaction reuses the normal system and tools", () =>
+  Effect.gen(function* () {
+    const { llm } = yield* useServerConfig(providerCfg)
+    const { prompt, chat } = yield* boot()
+    const compact = yield* SessionCompaction.Service
+    yield* user(chat.id, "cache prefix message")
+    yield* llm.text("normal response")
+    yield* prompt.loop({ sessionID: chat.id })
+
+    yield* compact.create({ sessionID: chat.id, agent: "build", model: ref, auto: false })
+    yield* llm.text("compacted summary")
+    yield* prompt.loop({ sessionID: chat.id })
+
+    const inputs = yield* llm.inputs
+    expect(inputs).toHaveLength(2)
+    expect(inputs[1]?.tools).toEqual(inputs[0]?.tools)
+    const normalMessages = Array.isArray(inputs[0]?.messages) ? inputs[0].messages : []
+    const compactMessages = Array.isArray(inputs[1]?.messages) ? inputs[1].messages : []
+    const system = (message: unknown) =>
+      typeof message === "object" && message !== null && "role" in message && message.role === "system"
+    expect(compactMessages.find(system)).toEqual(normalMessages.find(system))
+    expect(JSON.stringify(compactMessages)).toContain("cache prefix message")
+    expect(JSON.stringify(compactMessages)).toContain("normal response")
+    expect(JSON.stringify(compactMessages)).toContain("Create a new anchored summary")
+    expect(JSON.stringify(compactMessages)).not.toContain("[User]: cache prefix message")
+  }),
+)
+
 noLLMServer.instance(
  "loop exits immediately when last assistant has stop finish",
  () =>
@@ -86,7 +86,6 @@ OpenCode Zen هي بوابة AI تتيح لك الوصول إلى هذه الن
 | Claude Sonnet 4.5           | claude-sonnet-4-5           | `https://opencode.ai/zen/v1/messages`                     | `@ai-sdk/anthropic`         |
 | Claude Haiku 4.5            | claude-haiku-4-5            | `https://opencode.ai/zen/v1/messages`                     | `@ai-sdk/anthropic`         |
 | Gemini 3.6 Flash            | gemini-3.6-flash            | `https://opencode.ai/zen/v1/models/gemini-3.6-flash`      | `@ai-sdk/google`            |
-| Gemini 3.7 Flash            | gemini-3.7-flash            | `https://opencode.ai/zen/v1/models/gemini-3.7-flash`      | `@ai-sdk/google`            |
 | Gemini 3.5 Flash            | gemini-3.5-flash            | `https://opencode.ai/zen/v1/models/gemini-3.5-flash`      | `@ai-sdk/google`            |
 | Gemini 3.5 Flash Lite       | gemini-3.5-flash-lite       | `https://opencode.ai/zen/v1/models/gemini-3.5-flash-lite` | `@ai-sdk/google`            |
 | Gemini 3.1 Pro              | gemini-3.1-pro              | `https://opencode.ai/zen/v1/models/gemini-3.1-pro`        | `@ai-sdk/google`            |
@@ -173,7 +172,6 @@ https://opencode.ai/zen/v1/models
 | Claude Sonnet 4.5 (> 200K tokens) | $6.00   | $22.50  | $0.60           | $7.50           |
 | Claude Haiku 4.5                  | $1.00   | $5.00   | $0.10           | $1.25           |
 | Gemini 3.6 Flash                  | $1.50   | $7.50   | $0.15           | -               |
-| Gemini 3.7 Flash                  | $1.50   | $7.50   | $0.15           | -               |
 | Gemini 3.5 Flash                  | $1.50   | $9.00   | $0.15           | -               |
 | Gemini 3.5 Flash Lite             | $0.30   | $2.50   | $0.03           | -               |
 | Gemini 3.1 Pro (≤ 200K tokens)    | $2.00   | $12.00  | $0.20           | -               |
@@ -91,7 +91,6 @@ Našim modelima možete pristupiti i preko sljedećih API endpointa.
 | Claude Sonnet 4.5           | claude-sonnet-4-5           | `https://opencode.ai/zen/v1/messages`                     | `@ai-sdk/anthropic`         |
 | Claude Haiku 4.5            | claude-haiku-4-5            | `https://opencode.ai/zen/v1/messages`                     | `@ai-sdk/anthropic`         |
 | Gemini 3.6 Flash            | gemini-3.6-flash            | `https://opencode.ai/zen/v1/models/gemini-3.6-flash`      | `@ai-sdk/google`            |
-| Gemini 3.7 Flash            | gemini-3.7-flash            | `https://opencode.ai/zen/v1/models/gemini-3.7-flash`      | `@ai-sdk/google`            |
 | Gemini 3.5 Flash            | gemini-3.5-flash            | `https://opencode.ai/zen/v1/models/gemini-3.5-flash`      | `@ai-sdk/google`            |
 | Gemini 3.5 Flash Lite       | gemini-3.5-flash-lite       | `https://opencode.ai/zen/v1/models/gemini-3.5-flash-lite` | `@ai-sdk/google`            |
 | Gemini 3.1 Pro              | gemini-3.1-pro              | `https://opencode.ai/zen/v1/models/gemini-3.1-pro`        | `@ai-sdk/google`            |
@@ -180,7 +179,6 @@ Podržavamo pay-as-you-go model. Ispod su cijene **po 1M tokena**.
 | Claude Sonnet 4.5 (> 200K tokens) | $6.00  | $22.50  | $0.60       | $7.50        |
 | Claude Haiku 4.5                  | $1.00  | $5.00   | $0.10       | $1.25        |
 | Gemini 3.6 Flash                  | $1.50  | $7.50   | $0.15       | -            |
-| Gemini 3.7 Flash                  | $1.50  | $7.50   | $0.15       | -            |
 | Gemini 3.5 Flash                  | $1.50  | $9.00   | $0.15       | -            |
 | Gemini 3.5 Flash Lite             | $0.30  | $2.50   | $0.03       | -            |
 | Gemini 3.1 Pro (≤ 200K tokens)    | $2.00  | $12.00  | $0.20       | -            |
@@ -91,7 +91,6 @@ Du kan også få adgang til vores modeller gennem følgende API-endpoints.
 | Claude Sonnet 4.5           | claude-sonnet-4-5           | `https://opencode.ai/zen/v1/messages`                     | `@ai-sdk/anthropic`         |
 | Claude Haiku 4.5            | claude-haiku-4-5            | `https://opencode.ai/zen/v1/messages`                     | `@ai-sdk/anthropic`         |
 | Gemini 3.6 Flash            | gemini-3.6-flash            | `https://opencode.ai/zen/v1/models/gemini-3.6-flash`      | `@ai-sdk/google`            |
-| Gemini 3.7 Flash            | gemini-3.7-flash            | `https://opencode.ai/zen/v1/models/gemini-3.7-flash`      | `@ai-sdk/google`            |
 | Gemini 3.5 Flash            | gemini-3.5-flash            | `https://opencode.ai/zen/v1/models/gemini-3.5-flash`      | `@ai-sdk/google`            |
 | Gemini 3.5 Flash Lite       | gemini-3.5-flash-lite       | `https://opencode.ai/zen/v1/models/gemini-3.5-flash-lite` | `@ai-sdk/google`            |
 | Gemini 3.1 Pro              | gemini-3.1-pro              | `https://opencode.ai/zen/v1/models/gemini-3.1-pro`        | `@ai-sdk/google`            |
@@ -180,7 +179,6 @@ Vi understøtter en pay-as-you-go-model. Nedenfor er priserne **pr. 1M tokens**.
 | Claude Sonnet 4.5 (> 200K tokens) | $6.00  | $22.50  | $0.60       | $7.50        |
 | Claude Haiku 4.5                  | $1.00  | $5.00   | $0.10       | $1.25        |
 | Gemini 3.6 Flash                  | $1.50  | $7.50   | $0.15       | -            |
-| Gemini 3.7 Flash                  | $1.50  | $7.50   | $0.15       | -            |
 | Gemini 3.5 Flash                  | $1.50  | $9.00   | $0.15       | -            |
 | Gemini 3.5 Flash Lite             | $0.30  | $2.50   | $0.03       | -            |
 | Gemini 3.1 Pro (≤ 200K tokens)    | $2.00  | $12.00  | $0.20       | -            |
@@ -82,7 +82,6 @@ Du kannst auch über die folgenden API-Endpunkte auf unsere Modelle zugreifen.
 | Claude Sonnet 4.5           | claude-sonnet-4-5           | `https://opencode.ai/zen/v1/messages`                     | `@ai-sdk/anthropic`         |
 | Claude Haiku 4.5            | claude-haiku-4-5            | `https://opencode.ai/zen/v1/messages`                     | `@ai-sdk/anthropic`         |
 | Gemini 3.6 Flash            | gemini-3.6-flash            | `https://opencode.ai/zen/v1/models/gemini-3.6-flash`      | `@ai-sdk/google`            |
-| Gemini 3.7 Flash            | gemini-3.7-flash            | `https://opencode.ai/zen/v1/models/gemini-3.7-flash`      | `@ai-sdk/google`            |
 | Gemini 3.5 Flash            | gemini-3.5-flash            | `https://opencode.ai/zen/v1/models/gemini-3.5-flash`      | `@ai-sdk/google`            |
 | Gemini 3.5 Flash Lite       | gemini-3.5-flash-lite       | `https://opencode.ai/zen/v1/models/gemini-3.5-flash-lite` | `@ai-sdk/google`            |
 | Gemini 3.1 Pro              | gemini-3.1-pro              | `https://opencode.ai/zen/v1/models/gemini-3.1-pro`        | `@ai-sdk/google`            |
@@ -169,7 +168,6 @@ Wir unterstützen ein Pay-as-you-go-Modell. Unten findest du die Preise **pro 1M
 | Claude Sonnet 4.5 (> 200K tokens) | $6.00  | $22.50  | $0.60       | $7.50        |
 | Claude Haiku 4.5                  | $1.00  | $5.00   | $0.10       | $1.25        |
 | Gemini 3.6 Flash                  | $1.50  | $7.50   | $0.15       | -            |
-| Gemini 3.7 Flash                  | $1.50  | $7.50   | $0.15       | -            |
 | Gemini 3.5 Flash                  | $1.50  | $9.00   | $0.15       | -            |
 | Gemini 3.5 Flash Lite             | $0.30  | $2.50   | $0.03       | -            |
 | Gemini 3.1 Pro (≤ 200K tokens)    | $2.00  | $12.00  | $0.20       | -            |
@@ -91,7 +91,6 @@ También puedes acceder a nuestros modelos a través de los siguientes endpoints
 | Claude Sonnet 4.5           | claude-sonnet-4-5           | `https://opencode.ai/zen/v1/messages`                     | `@ai-sdk/anthropic`         |
 | Claude Haiku 4.5            | claude-haiku-4-5            | `https://opencode.ai/zen/v1/messages`                     | `@ai-sdk/anthropic`         |
 | Gemini 3.6 Flash            | gemini-3.6-flash            | `https://opencode.ai/zen/v1/models/gemini-3.6-flash`      | `@ai-sdk/google`            |
-| Gemini 3.7 Flash            | gemini-3.7-flash            | `https://opencode.ai/zen/v1/models/gemini-3.7-flash`      | `@ai-sdk/google`            |
 | Gemini 3.5 Flash            | gemini-3.5-flash            | `https://opencode.ai/zen/v1/models/gemini-3.5-flash`      | `@ai-sdk/google`            |
 | Gemini 3.5 Flash Lite       | gemini-3.5-flash-lite       | `https://opencode.ai/zen/v1/models/gemini-3.5-flash-lite` | `@ai-sdk/google`            |
 | Gemini 3.1 Pro              | gemini-3.1-pro              | `https://opencode.ai/zen/v1/models/gemini-3.1-pro`        | `@ai-sdk/google`            |
@@ -180,7 +179,6 @@ Admitimos un modelo de pago por uso. A continuación se muestran los precios **p
 | Claude Sonnet 4.5 (> 200K tokens) | $6.00   | $22.50  | $0.60            | $7.50              |
 | Claude Haiku 4.5                  | $1.00   | $5.00   | $0.10            | $1.25              |
 | Gemini 3.6 Flash                  | $1.50   | $7.50   | $0.15            | -                  |
-| Gemini 3.7 Flash                  | $1.50   | $7.50   | $0.15            | -                  |
 | Gemini 3.5 Flash                  | $1.50   | $9.00   | $0.15            | -                  |
 | Gemini 3.5 Flash Lite             | $0.30   | $2.50   | $0.03            | -                  |
 | Gemini 3.1 Pro (≤ 200K tokens)    | $2.00   | $12.00  | $0.20            | -                  |
@@ -82,7 +82,6 @@ Vous pouvez également accéder à nos modèles via les points de terminaison AP
 | Claude Sonnet 4.5           | claude-sonnet-4-5           | `https://opencode.ai/zen/v1/messages`                     | `@ai-sdk/anthropic`         |
 | Claude Haiku 4.5            | claude-haiku-4-5            | `https://opencode.ai/zen/v1/messages`                     | `@ai-sdk/anthropic`         |
 | Gemini 3.6 Flash            | gemini-3.6-flash            | `https://opencode.ai/zen/v1/models/gemini-3.6-flash`      | `@ai-sdk/google`            |
-| Gemini 3.7 Flash            | gemini-3.7-flash            | `https://opencode.ai/zen/v1/models/gemini-3.7-flash`      | `@ai-sdk/google`            |
 | Gemini 3.5 Flash            | gemini-3.5-flash            | `https://opencode.ai/zen/v1/models/gemini-3.5-flash`      | `@ai-sdk/google`            |
 | Gemini 3.5 Flash Lite       | gemini-3.5-flash-lite       | `https://opencode.ai/zen/v1/models/gemini-3.5-flash-lite` | `@ai-sdk/google`            |
 | Gemini 3.1 Pro              | gemini-3.1-pro              | `https://opencode.ai/zen/v1/models/gemini-3.1-pro`        | `@ai-sdk/google`            |
@@ -169,7 +168,6 @@ Nous prenons en charge un modèle de paiement à l'utilisation. Vous trouverez c
 | Claude Sonnet 4.5 (> 200K tokens) | $6.00  | $22.50  | $0.60       | $7.50        |
 | Claude Haiku 4.5                  | $1.00  | $5.00   | $0.10       | $1.25        |
 | Gemini 3.6 Flash                  | $1.50  | $7.50   | $0.15       | -            |
-| Gemini 3.7 Flash                  | $1.50  | $7.50   | $0.15       | -            |
 | Gemini 3.5 Flash                  | $1.50  | $9.00   | $0.15       | -            |
 | Gemini 3.5 Flash Lite             | $0.30  | $2.50   | $0.03       | -            |
 | Gemini 3.1 Pro (≤ 200K tokens)    | $2.00  | $12.00  | $0.20       | -            |
@@ -91,7 +91,6 @@ Puoi anche accedere ai nostri modelli tramite i seguenti endpoint API.
 | Claude Sonnet 4.5           | claude-sonnet-4-5           | `https://opencode.ai/zen/v1/messages`                     | `@ai-sdk/anthropic`         |
 | Claude Haiku 4.5            | claude-haiku-4-5            | `https://opencode.ai/zen/v1/messages`                     | `@ai-sdk/anthropic`         |
 | Gemini 3.6 Flash            | gemini-3.6-flash            | `https://opencode.ai/zen/v1/models/gemini-3.6-flash`      | `@ai-sdk/google`            |
-| Gemini 3.7 Flash            | gemini-3.7-flash            | `https://opencode.ai/zen/v1/models/gemini-3.7-flash`      | `@ai-sdk/google`            |
 | Gemini 3.5 Flash            | gemini-3.5-flash            | `https://opencode.ai/zen/v1/models/gemini-3.5-flash`      | `@ai-sdk/google`            |
 | Gemini 3.5 Flash Lite       | gemini-3.5-flash-lite       | `https://opencode.ai/zen/v1/models/gemini-3.5-flash-lite` | `@ai-sdk/google`            |
 | Gemini 3.1 Pro              | gemini-3.1-pro              | `https://opencode.ai/zen/v1/models/gemini-3.1-pro`        | `@ai-sdk/google`            |
@@ -180,7 +179,6 @@ Supportiamo un modello pay-as-you-go. Qui sotto trovi i prezzi **per 1M token**.
 | Claude Sonnet 4.5 (> 200K tokens) | $6.00  | $22.50  | $0.60       | $7.50        |
 | Claude Haiku 4.5                  | $1.00  | $5.00   | $0.10       | $1.25        |
 | Gemini 3.6 Flash                  | $1.50  | $7.50   | $0.15       | -            |
-| Gemini 3.7 Flash                  | $1.50  | $7.50   | $0.15       | -            |
 | Gemini 3.5 Flash                  | $1.50  | $9.00   | $0.15       | -            |
 | Gemini 3.5 Flash Lite             | $0.30  | $2.50   | $0.03       | -            |
 | Gemini 3.1 Pro (≤ 200K tokens)    | $2.00  | $12.00  | $0.20       | -            |
@@ -82,7 +82,6 @@ OpenCode Zen は、OpenCode のほかのプロバイダーと同じように動
 | Claude Sonnet 4.5           | claude-sonnet-4-5           | `https://opencode.ai/zen/v1/messages`                     | `@ai-sdk/anthropic`         |
 | Claude Haiku 4.5            | claude-haiku-4-5            | `https://opencode.ai/zen/v1/messages`                     | `@ai-sdk/anthropic`         |
 | Gemini 3.6 Flash            | gemini-3.6-flash            | `https://opencode.ai/zen/v1/models/gemini-3.6-flash`      | `@ai-sdk/google`            |
-| Gemini 3.7 Flash            | gemini-3.7-flash            | `https://opencode.ai/zen/v1/models/gemini-3.7-flash`      | `@ai-sdk/google`            |
 | Gemini 3.5 Flash            | gemini-3.5-flash            | `https://opencode.ai/zen/v1/models/gemini-3.5-flash`      | `@ai-sdk/google`            |
 | Gemini 3.5 Flash Lite       | gemini-3.5-flash-lite       | `https://opencode.ai/zen/v1/models/gemini-3.5-flash-lite` | `@ai-sdk/google`            |
 | Gemini 3.1 Pro              | gemini-3.1-pro              | `https://opencode.ai/zen/v1/models/gemini-3.1-pro`        | `@ai-sdk/google`            |
@@ -169,7 +168,6 @@ https://opencode.ai/zen/v1/models
 | Claude Sonnet 4.5 (> 200K tokens) | $6.00  | $22.50  | $0.60       | $7.50        |
 | Claude Haiku 4.5                  | $1.00  | $5.00   | $0.10       | $1.25        |
 | Gemini 3.6 Flash                  | $1.50  | $7.50   | $0.15       | -            |
-| Gemini 3.7 Flash                  | $1.50  | $7.50   | $0.15       | -            |
 | Gemini 3.5 Flash                  | $1.50  | $9.00   | $0.15       | -            |
 | Gemini 3.5 Flash Lite             | $0.30  | $2.50   | $0.03       | -            |
 | Gemini 3.1 Pro (≤ 200K tokens)    | $2.00  | $12.00  | $0.20       | -            |
@@ -82,7 +82,6 @@ OpenCode Zen은 OpenCode의 다른 provider와 똑같이 작동합니다.
 | Claude Sonnet 4.5           | claude-sonnet-4-5           | `https://opencode.ai/zen/v1/messages`                     | `@ai-sdk/anthropic`         |
 | Claude Haiku 4.5            | claude-haiku-4-5            | `https://opencode.ai/zen/v1/messages`                     | `@ai-sdk/anthropic`         |
 | Gemini 3.6 Flash            | gemini-3.6-flash            | `https://opencode.ai/zen/v1/models/gemini-3.6-flash`      | `@ai-sdk/google`            |
-| Gemini 3.7 Flash            | gemini-3.7-flash            | `https://opencode.ai/zen/v1/models/gemini-3.7-flash`      | `@ai-sdk/google`            |
 | Gemini 3.5 Flash            | gemini-3.5-flash            | `https://opencode.ai/zen/v1/models/gemini-3.5-flash`      | `@ai-sdk/google`            |
 | Gemini 3.5 Flash Lite       | gemini-3.5-flash-lite       | `https://opencode.ai/zen/v1/models/gemini-3.5-flash-lite` | `@ai-sdk/google`            |
 | Gemini 3.1 Pro              | gemini-3.1-pro              | `https://opencode.ai/zen/v1/models/gemini-3.1-pro`        | `@ai-sdk/google`            |
@@ -169,7 +168,6 @@ https://opencode.ai/zen/v1/models
 | Claude Sonnet 4.5 (> 200K tokens) | $6.00  | $22.50  | $0.60       | $7.50        |
 | Claude Haiku 4.5                  | $1.00  | $5.00   | $0.10       | $1.25        |
 | Gemini 3.6 Flash                  | $1.50  | $7.50   | $0.15       | -            |
-| Gemini 3.7 Flash                  | $1.50  | $7.50   | $0.15       | -            |
 | Gemini 3.5 Flash                  | $1.50  | $9.00   | $0.15       | -            |
 | Gemini 3.5 Flash Lite             | $0.30  | $2.50   | $0.03       | -            |
 | Gemini 3.1 Pro (≤ 200K tokens)    | $2.00  | $12.00  | $0.20       | -            |
@@ -91,7 +91,6 @@ Du kan også få tilgang til modellene våre gjennom følgende API-endepunkter.
 | Claude Sonnet 4.5           | claude-sonnet-4-5           | `https://opencode.ai/zen/v1/messages`                     | `@ai-sdk/anthropic`         |
 | Claude Haiku 4.5            | claude-haiku-4-5            | `https://opencode.ai/zen/v1/messages`                     | `@ai-sdk/anthropic`         |
 | Gemini 3.6 Flash            | gemini-3.6-flash            | `https://opencode.ai/zen/v1/models/gemini-3.6-flash`      | `@ai-sdk/google`            |
-| Gemini 3.7 Flash            | gemini-3.7-flash            | `https://opencode.ai/zen/v1/models/gemini-3.7-flash`      | `@ai-sdk/google`            |
 | Gemini 3.5 Flash            | gemini-3.5-flash            | `https://opencode.ai/zen/v1/models/gemini-3.5-flash`      | `@ai-sdk/google`            |
 | Gemini 3.5 Flash Lite       | gemini-3.5-flash-lite       | `https://opencode.ai/zen/v1/models/gemini-3.5-flash-lite` | `@ai-sdk/google`            |
 | Gemini 3.1 Pro              | gemini-3.1-pro              | `https://opencode.ai/zen/v1/models/gemini-3.1-pro`        | `@ai-sdk/google`            |
@@ -180,7 +179,6 @@ Vi støtter en pay-as-you-go-modell. Nedenfor er prisene **per 1M tokens**.
 | Claude Sonnet 4.5 (> 200K tokens) | $6.00   | $22.50  | $0.60         | $7.50           |
 | Claude Haiku 4.5                  | $1.00   | $5.00   | $0.10         | $1.25           |
 | Gemini 3.6 Flash                  | $1.50   | $7.50   | $0.15         | -               |
-| Gemini 3.7 Flash                  | $1.50   | $7.50   | $0.15         | -               |
 | Gemini 3.5 Flash                  | $1.50   | $9.00   | $0.15         | -               |
 | Gemini 3.5 Flash Lite             | $0.30   | $2.50   | $0.03         | -               |
 | Gemini 3.1 Pro (≤ 200K tokens)    | $2.00   | $12.00  | $0.20         | -               |
@@ -91,7 +91,6 @@ Możesz też uzyskać dostęp do naszych modeli przez poniższe endpointy API.
 | Claude Sonnet 4.5           | claude-sonnet-4-5           | `https://opencode.ai/zen/v1/messages`                     | `@ai-sdk/anthropic`         |
 | Claude Haiku 4.5            | claude-haiku-4-5            | `https://opencode.ai/zen/v1/messages`                     | `@ai-sdk/anthropic`         |
 | Gemini 3.6 Flash            | gemini-3.6-flash            | `https://opencode.ai/zen/v1/models/gemini-3.6-flash`      | `@ai-sdk/google`            |
-| Gemini 3.7 Flash            | gemini-3.7-flash            | `https://opencode.ai/zen/v1/models/gemini-3.7-flash`      | `@ai-sdk/google`            |
 | Gemini 3.5 Flash            | gemini-3.5-flash            | `https://opencode.ai/zen/v1/models/gemini-3.5-flash`      | `@ai-sdk/google`            |
 | Gemini 3.5 Flash Lite       | gemini-3.5-flash-lite       | `https://opencode.ai/zen/v1/models/gemini-3.5-flash-lite` | `@ai-sdk/google`            |
 | Gemini 3.1 Pro              | gemini-3.1-pro              | `https://opencode.ai/zen/v1/models/gemini-3.1-pro`        | `@ai-sdk/google`            |
@@ -180,7 +179,6 @@ Obsługujemy model pay-as-you-go. Poniżej znajdują się ceny **za 1M tokenów*
 | Claude Sonnet 4.5 (> 200K tokens) | $6.00   | $22.50  | $0.60          | $7.50          |
 | Claude Haiku 4.5                  | $1.00   | $5.00   | $0.10          | $1.25          |
 | Gemini 3.6 Flash                  | $1.50   | $7.50   | $0.15          | -              |
-| Gemini 3.7 Flash                  | $1.50   | $7.50   | $0.15          | -              |
 | Gemini 3.5 Flash                  | $1.50   | $9.00   | $0.15          | -              |
 | Gemini 3.5 Flash Lite             | $0.30   | $2.50   | $0.03          | -              |
 | Gemini 3.1 Pro (≤ 200K tokens)    | $2.00   | $12.00  | $0.20          | -              |
@@ -1718,10 +1718,6 @@ We recommend signing up for [ChatGPT Plus or Pro](https://chatgpt.com/pricing).
   /models
   ```

-##### Compute residency
-
-For ChatGPT OAuth, OpenCode automatically applies a regional inference residency requirement when one is advertised by your workspace credentials. It forwards the compute residency value from the credential instead of maintaining a fixed list of regions. Data residency at rest does not imply regional inference.
-
 ##### Using API keys

 If you already have an API key, you can select **Manually enter API Key** and paste it in your terminal.
@@ -82,7 +82,6 @@ Você também pode acessar nossos modelos pelos seguintes endpoints de API.
 | Claude Sonnet 4.5           | claude-sonnet-4-5           | `https://opencode.ai/zen/v1/messages`                     | `@ai-sdk/anthropic`         |
 | Claude Haiku 4.5            | claude-haiku-4-5            | `https://opencode.ai/zen/v1/messages`                     | `@ai-sdk/anthropic`         |
 | Gemini 3.6 Flash            | gemini-3.6-flash            | `https://opencode.ai/zen/v1/models/gemini-3.6-flash`      | `@ai-sdk/google`            |
-| Gemini 3.7 Flash            | gemini-3.7-flash            | `https://opencode.ai/zen/v1/models/gemini-3.7-flash`      | `@ai-sdk/google`            |
 | Gemini 3.5 Flash            | gemini-3.5-flash            | `https://opencode.ai/zen/v1/models/gemini-3.5-flash`      | `@ai-sdk/google`            |
 | Gemini 3.5 Flash Lite       | gemini-3.5-flash-lite       | `https://opencode.ai/zen/v1/models/gemini-3.5-flash-lite` | `@ai-sdk/google`            |
 | Gemini 3.1 Pro              | gemini-3.1-pro              | `https://opencode.ai/zen/v1/models/gemini-3.1-pro`        | `@ai-sdk/google`            |
@@ -169,7 +168,6 @@ Oferecemos um modelo pay-as-you-go. Abaixo estão os preços **por 1M tokens**.
 | Claude Sonnet 4.5 (> 200K tokens) | $6.00   | $22.50  | $0.60            | $7.50            |
 | Claude Haiku 4.5                  | $1.00   | $5.00   | $0.10            | $1.25            |
 | Gemini 3.6 Flash                  | $1.50   | $7.50   | $0.15            | -                |
-| Gemini 3.7 Flash                  | $1.50   | $7.50   | $0.15            | -                |
 | Gemini 3.5 Flash                  | $1.50   | $9.00   | $0.15            | -                |
 | Gemini 3.5 Flash Lite             | $0.30   | $2.50   | $0.03            | -                |
 | Gemini 3.1 Pro (≤ 200K tokens)    | $2.00   | $12.00  | $0.20            | -                |
@@ -91,7 +91,6 @@ OpenCode Zen работает как любой другой провайдер
 | Claude Sonnet 4.5           | claude-sonnet-4-5           | `https://opencode.ai/zen/v1/messages`                     | `@ai-sdk/anthropic`         |
 | Claude Haiku 4.5            | claude-haiku-4-5            | `https://opencode.ai/zen/v1/messages`                     | `@ai-sdk/anthropic`         |
 | Gemini 3.6 Flash            | gemini-3.6-flash            | `https://opencode.ai/zen/v1/models/gemini-3.6-flash`      | `@ai-sdk/google`            |
-| Gemini 3.7 Flash            | gemini-3.7-flash            | `https://opencode.ai/zen/v1/models/gemini-3.7-flash`      | `@ai-sdk/google`            |
 | Gemini 3.5 Flash            | gemini-3.5-flash            | `https://opencode.ai/zen/v1/models/gemini-3.5-flash`      | `@ai-sdk/google`            |
 | Gemini 3.5 Flash Lite       | gemini-3.5-flash-lite       | `https://opencode.ai/zen/v1/models/gemini-3.5-flash-lite` | `@ai-sdk/google`            |
 | Gemini 3.1 Pro              | gemini-3.1-pro              | `https://opencode.ai/zen/v1/models/gemini-3.1-pro`        | `@ai-sdk/google`            |
@@ -180,7 +179,6 @@ https://opencode.ai/zen/v1/models
 | Claude Sonnet 4.5 (> 200K tokens) | $6.00  | $22.50  | $0.60       | $7.50        |
 | Claude Haiku 4.5                  | $1.00  | $5.00   | $0.10       | $1.25        |
 | Gemini 3.6 Flash                  | $1.50  | $7.50   | $0.15       | -            |
-| Gemini 3.7 Flash                  | $1.50  | $7.50   | $0.15       | -            |
 | Gemini 3.5 Flash                  | $1.50  | $9.00   | $0.15       | -            |
 | Gemini 3.5 Flash Lite             | $0.30  | $2.50   | $0.03       | -            |
 | Gemini 3.1 Pro (≤ 200K tokens)    | $2.00  | $12.00  | $0.20       | -            |
@@ -84,7 +84,6 @@ OpenCode Zen ทำงานเหมือน provider อื่น ๆ ใน
 | Claude Sonnet 4.5           | claude-sonnet-4-5           | `https://opencode.ai/zen/v1/messages`                     | `@ai-sdk/anthropic`         |
 | Claude Haiku 4.5            | claude-haiku-4-5            | `https://opencode.ai/zen/v1/messages`                     | `@ai-sdk/anthropic`         |
 | Gemini 3.6 Flash            | gemini-3.6-flash            | `https://opencode.ai/zen/v1/models/gemini-3.6-flash`      | `@ai-sdk/google`            |
-| Gemini 3.7 Flash            | gemini-3.7-flash            | `https://opencode.ai/zen/v1/models/gemini-3.7-flash`      | `@ai-sdk/google`            |
 | Gemini 3.5 Flash            | gemini-3.5-flash            | `https://opencode.ai/zen/v1/models/gemini-3.5-flash`      | `@ai-sdk/google`            |
 | Gemini 3.5 Flash Lite       | gemini-3.5-flash-lite       | `https://opencode.ai/zen/v1/models/gemini-3.5-flash-lite` | `@ai-sdk/google`            |
 | Gemini 3.1 Pro              | gemini-3.1-pro              | `https://opencode.ai/zen/v1/models/gemini-3.1-pro`        | `@ai-sdk/google`            |
@@ -171,7 +170,6 @@ https://opencode.ai/zen/v1/models
 | Claude Sonnet 4.5 (> 200K tokens) | $6.00  | $22.50  | $0.60       | $7.50        |
 | Claude Haiku 4.5                  | $1.00  | $5.00   | $0.10       | $1.25        |
 | Gemini 3.6 Flash                  | $1.50  | $7.50   | $0.15       | -            |
-| Gemini 3.7 Flash                  | $1.50  | $7.50   | $0.15       | -            |
 | Gemini 3.5 Flash                  | $1.50  | $9.00   | $0.15       | -            |
 | Gemini 3.5 Flash Lite             | $0.30  | $2.50   | $0.03       | -            |
 | Gemini 3.1 Pro (≤ 200K tokens)    | $2.00  | $12.00  | $0.20       | -            |
@@ -82,7 +82,6 @@ Modellerimize aşağıdaki API uç noktaları aracılığıyla da erişebilirsin
 | Claude Sonnet 4.5           | claude-sonnet-4-5           | `https://opencode.ai/zen/v1/messages`                     | `@ai-sdk/anthropic`         |
 | Claude Haiku 4.5            | claude-haiku-4-5            | `https://opencode.ai/zen/v1/messages`                     | `@ai-sdk/anthropic`         |
 | Gemini 3.6 Flash            | gemini-3.6-flash            | `https://opencode.ai/zen/v1/models/gemini-3.6-flash`      | `@ai-sdk/google`            |
-| Gemini 3.7 Flash            | gemini-3.7-flash            | `https://opencode.ai/zen/v1/models/gemini-3.7-flash`      | `@ai-sdk/google`            |
 | Gemini 3.5 Flash            | gemini-3.5-flash            | `https://opencode.ai/zen/v1/models/gemini-3.5-flash`      | `@ai-sdk/google`            |
 | Gemini 3.5 Flash Lite       | gemini-3.5-flash-lite       | `https://opencode.ai/zen/v1/models/gemini-3.5-flash-lite` | `@ai-sdk/google`            |
 | Gemini 3.1 Pro              | gemini-3.1-pro              | `https://opencode.ai/zen/v1/models/gemini-3.1-pro`        | `@ai-sdk/google`            |
@@ -169,7 +168,6 @@ Kullandıkça öde modelini destekliyoruz. Aşağıda **1M token başına** fiya
 | Claude Sonnet 4.5 (> 200K tokens) | $6.00  | $22.50  | $0.60       | $7.50        |
 | Claude Haiku 4.5                  | $1.00  | $5.00   | $0.10       | $1.25        |
 | Gemini 3.6 Flash                  | $1.50  | $7.50   | $0.15       | -            |
-| Gemini 3.7 Flash                  | $1.50  | $7.50   | $0.15       | -            |
 | Gemini 3.5 Flash                  | $1.50  | $9.00   | $0.15       | -            |
 | Gemini 3.5 Flash Lite             | $0.30  | $2.50   | $0.03       | -            |
 | Gemini 3.1 Pro (≤ 200K tokens)    | $2.00  | $12.00  | $0.20       | -            |
@@ -91,7 +91,6 @@ You can also access our models through the following API endpoints.
 | Claude Sonnet 4.5           | claude-sonnet-4-5           | `https://opencode.ai/zen/v1/messages`                     | `@ai-sdk/anthropic`         |
 | Claude Haiku 4.5            | claude-haiku-4-5            | `https://opencode.ai/zen/v1/messages`                     | `@ai-sdk/anthropic`         |
 | Gemini 3.6 Flash            | gemini-3.6-flash            | `https://opencode.ai/zen/v1/models/gemini-3.6-flash`      | `@ai-sdk/google`            |
-| Gemini 3.7 Flash            | gemini-3.7-flash            | `https://opencode.ai/zen/v1/models/gemini-3.7-flash`      | `@ai-sdk/google`            |
 | Gemini 3.5 Flash            | gemini-3.5-flash            | `https://opencode.ai/zen/v1/models/gemini-3.5-flash`      | `@ai-sdk/google`            |
 | Gemini 3.5 Flash Lite       | gemini-3.5-flash-lite       | `https://opencode.ai/zen/v1/models/gemini-3.5-flash-lite` | `@ai-sdk/google`            |
 | Gemini 3.1 Pro              | gemini-3.1-pro              | `https://opencode.ai/zen/v1/models/gemini-3.1-pro`        | `@ai-sdk/google`            |
@@ -180,7 +179,6 @@ We support a pay-as-you-go model. Below are the prices **per 1M tokens**.
 | Claude Sonnet 4.5 (> 200K tokens) | $6.00  | $22.50  | $0.60       | $7.50        |
 | Claude Haiku 4.5                  | $1.00  | $5.00   | $0.10       | $1.25        |
 | Gemini 3.6 Flash                  | $1.50  | $7.50   | $0.15       | -            |
-| Gemini 3.7 Flash                  | $1.50  | $7.50   | $0.15       | -            |
 | Gemini 3.5 Flash                  | $1.50  | $9.00   | $0.15       | -            |
 | Gemini 3.5 Flash Lite             | $0.30  | $2.50   | $0.03       | -            |
 | Gemini 3.1 Pro (≤ 200K tokens)    | $2.00  | $12.00  | $0.20       | -            |
@@ -82,7 +82,6 @@ OpenCode Zen 的工作方式与 OpenCode 中的任何其他提供商相同。
 | Claude Sonnet 4.5           | claude-sonnet-4-5           | `https://opencode.ai/zen/v1/messages`                     | `@ai-sdk/anthropic`         |
 | Claude Haiku 4.5            | claude-haiku-4-5            | `https://opencode.ai/zen/v1/messages`                     | `@ai-sdk/anthropic`         |
 | Gemini 3.6 Flash            | gemini-3.6-flash            | `https://opencode.ai/zen/v1/models/gemini-3.6-flash`      | `@ai-sdk/google`            |
-| Gemini 3.7 Flash            | gemini-3.7-flash            | `https://opencode.ai/zen/v1/models/gemini-3.7-flash`      | `@ai-sdk/google`            |
 | Gemini 3.5 Flash            | gemini-3.5-flash            | `https://opencode.ai/zen/v1/models/gemini-3.5-flash`      | `@ai-sdk/google`            |
 | Gemini 3.5 Flash Lite       | gemini-3.5-flash-lite       | `https://opencode.ai/zen/v1/models/gemini-3.5-flash-lite` | `@ai-sdk/google`            |
 | Gemini 3.1 Pro              | gemini-3.1-pro              | `https://opencode.ai/zen/v1/models/gemini-3.1-pro`        | `@ai-sdk/google`            |
@@ -169,7 +168,6 @@ https://opencode.ai/zen/v1/models
 | Claude Sonnet 4.5 (> 200K tokens) | $6.00  | $22.50  | $0.60    | $7.50    |
 | Claude Haiku 4.5                  | $1.00  | $5.00   | $0.10    | $1.25    |
 | Gemini 3.6 Flash                  | $1.50  | $7.50   | $0.15    | -        |
-| Gemini 3.7 Flash                  | $1.50  | $7.50   | $0.15    | -        |
 | Gemini 3.5 Flash                  | $1.50  | $9.00   | $0.15    | -        |
 | Gemini 3.5 Flash Lite             | $0.30  | $2.50   | $0.03    | -        |
 | Gemini 3.1 Pro (≤ 200K tokens)    | $2.00  | $12.00  | $0.20    | -        |
@@ -86,7 +86,6 @@ OpenCode Zen 的運作方式和 OpenCode 中的其他供應商一樣。
 | Claude Sonnet 4.5           | claude-sonnet-4-5           | `https://opencode.ai/zen/v1/messages`                     | `@ai-sdk/anthropic`         |
 | Claude Haiku 4.5            | claude-haiku-4-5            | `https://opencode.ai/zen/v1/messages`                     | `@ai-sdk/anthropic`         |
 | Gemini 3.6 Flash            | gemini-3.6-flash            | `https://opencode.ai/zen/v1/models/gemini-3.6-flash`      | `@ai-sdk/google`            |
-| Gemini 3.7 Flash            | gemini-3.7-flash            | `https://opencode.ai/zen/v1/models/gemini-3.7-flash`      | `@ai-sdk/google`            |
 | Gemini 3.5 Flash            | gemini-3.5-flash            | `https://opencode.ai/zen/v1/models/gemini-3.5-flash`      | `@ai-sdk/google`            |
 | Gemini 3.5 Flash Lite       | gemini-3.5-flash-lite       | `https://opencode.ai/zen/v1/models/gemini-3.5-flash-lite` | `@ai-sdk/google`            |
 | Gemini 3.1 Pro              | gemini-3.1-pro              | `https://opencode.ai/zen/v1/models/gemini-3.1-pro`        | `@ai-sdk/google`            |
@@ -174,7 +173,6 @@ https://opencode.ai/zen/v1/models
 | Claude Sonnet 4.5 (> 200K tokens) | $6.00  | $22.50  | $0.60    | $7.50    |
 | Claude Haiku 4.5                  | $1.00  | $5.00   | $0.10    | $1.25    |
 | Gemini 3.6 Flash                  | $1.50  | $7.50   | $0.15    | -        |
-| Gemini 3.7 Flash                  | $1.50  | $7.50   | $0.15    | -        |
 | Gemini 3.5 Flash                  | $1.50  | $9.00   | $0.15    | -        |
 | Gemini 3.5 Flash Lite             | $0.30  | $2.50   | $0.03    | -        |
 | Gemini 3.1 Pro (≤ 200K tokens)    | $2.00  | $12.00  | $0.20    | -        |