fix(web-search): strip Lite marker from classic search

docs-site/src/content/docs/reference/configuration/providers.md

@@ -370,7 +370,7 @@ Providers can expose a built-in shorthand, such as `agy` for `google-antigravity
 | `defaultMaxOutputTokens?` | `number` | Provider-wide `openai-chat` fallback when the client omits `max_output_tokens`. |
 | `modelMaxOutputTokens?` | `Record<string, number>` | Positive per-model `openai-chat` fallback budgets; exact/pattern matches beat the provider default. |
 | `modelCosts?` | `Record<string, ProviderCostOverlay>` | Per-model display prices (USD per 1M tokens), keyed by that provider's exact upstream model id — not a provider identifier or a routed `provider/model` label, e.g. `{ "deepseek-v4-flash": { "input": 0.14, "output": 0.28, "cacheRead": 0.0028, "cacheWrite": 0 } }`. Any model id is a valid key — custom providers may target any OpenAI-compatible endpoint through the `openai-chat` adapter, and local or internal provider ids work even when they are absent from the built-in catalogs. User-configured prices win over the built-in catalogs in the Logs `~$` and Usage estimates; historical entries are repriced from the current overlay, so editing a price can move past totals. The fallback order is user `modelCosts` → exact official correction → jawcode catalog → expected-price overlay → model-level vendor fallback, and an explicit all-zero user entry means a known-zero estimate; delete that model entry to restore automatic pricing. All-zero catalog metadata still falls through. Each rate must be a non-negative finite number at most 1,000,000 (USD per 1M tokens); out-of-range rows are rejected by the management boundary and dropped on load. Display-time estimation only: overlays never affect routing, account selection, quotas, or billing. |
-| `headers?` | `Record<string, string>` | Extra upstream headers. Authorization, cookies, API-key headers, embedded newlines, and invalid names are rejected. Requests that ride the provider outbound wrapper — model discovery, connection tests, and other proxy-originated diagnostics such as the Ollama show probe — send `User-Agent: opencodex` unless a User-Agent is set here or by the provider preset. An experimental relay may negotiate earlier completed-tool delivery with `x-opencodex-tool-completion: block-stop-v1`; the embedding shell must supply authenticated durable turn membership, and the relay must acknowledge the same response header and implement exact call-bound completion markers. Leave it unset for other providers. Early delivery applies only to eligible structured calls; later errors or cancellation cannot retract a call already delivered. |
+| `headers?` | `Record<string, string>` | Extra upstream headers. For the classic OpenAI web-search sidecar, `x-openai-internal-codex-responses-lite` is removed case-insensitively after configured and selected headers are merged; other forwarded headers are preserved. Authorization, cookies, API-key headers, embedded newlines, and invalid names are rejected. Requests that ride the provider outbound wrapper — model discovery, connection tests, and other proxy-originated diagnostics such as the Ollama show probe — send `User-Agent: opencodex` unless a User-Agent is set here or by the provider preset. An experimental relay may negotiate earlier completed-tool delivery with `x-opencodex-tool-completion: block-stop-v1`; the embedding shell must supply authenticated durable turn membership, and the relay must acknowledge the same response header and implement exact call-bound completion markers. Leave it unset for other providers. Early delivery applies only to eligible structured calls; later errors or cancellation cannot retract a call already delivered. |
 | `forwardClientHeaders?` | `string[]` | `openai-responses` only: opt in to copying named inbound client metadata headers onto inference requests when the provider has not already set that header. Credential and transport-owned names such as `Authorization`, cookies, API-key headers, `Content-Type`, `Content-Length`, `Host`, and `x-oai-attestation` are rejected and also filtered at runtime. Header names are case-insensitive and may not repeat. `headers` remains authoritative; the existing caller `User-Agent` fallback still applies when neither surface sets one.  Supported names: `originator`, `x-client-request-id`, `x-codex-app-version`, `user-agent`; all other names are refused. |
 | `supportsNativeCompactionTrigger?` | `boolean` | `openai-responses` only: declare support for native v2 `compaction_trigger` input items. Defaults off for noncanonical gateways; preserves native compaction rather than requesting a portable text summary. Independent of the v1 endpoint and blob replay capabilities. |
 | `supportsNativeCompactEndpoint?` | `boolean` | `openai-responses` only: declare support for native v1 `POST /responses/compact`. Defaults off for custom gateways. Key/local/forward routes use the provider's Responses URL plus `/compact` and its normal transport authentication; OAuth keeps routed compaction and its existing credential lifecycle. Does not enable v2 triggers or grant official OpenAI identity. |

src/web-search/executor.ts

@@ -1,5 +1,5 @@
 import type { OcxProviderConfig } from "../types";
-import { mergeCodexSelectedHeaders } from "../codex/forward-transport-headers";
+import { CODEX_RESPONSES_LITE_HEADER, mergeCodexSelectedHeaders } from "../codex/forward-transport-headers";
 import { signalWithTimeout, cancelBodyOnAbort } from "../lib/abort";
 import { redactSecretString } from "../lib/redact";
 import { sidecarEnter } from "../lib/sidecar-tracker";
@@ -84,9 +84,11 @@ export async function runWebSearch(
   if (settings.reserveCompatibility && settings.model === NATIVE_RESERVE_MODEL) {
     return { text: "", sources: [], error: "Luna Reserve compatibility is only available as a conversation model, not a search helper. Choose another search helper model." };
   }
-  const headers = Object.fromEntries(mergeCodexSelectedHeaders(
+  const mergedHeaders = mergeCodexSelectedHeaders(
     { "Content-Type": "application/json", ...forwardProvider.headers }, selectedForwardHeaders,
-  ));
+  );
+  mergedHeaders.delete(CODEX_RESPONSES_LITE_HEADER);
+  const headers = Object.fromEntries(mergedHeaders);
   const body = {
     model: settings.model,
     instructions: settings.describeImages ? BASE_INSTRUCTION + IMAGE_INSTRUCTION : BASE_INSTRUCTION,

structure/providers-and-adapters.md

@@ -499,7 +499,7 @@ request-scoped OpenAI sidecar authority on completion, failure, and client cance
 cancellation releases immediately rather than waiting on an abandoned upstream read — so a
 recovery probe lease no search consumed is always returned.
 `tests/web-search/web-search-bridge-replay.test.ts` pins the restore and each of those refusals.
-A forward OpenAI search sidecar retries a 429 only when the requested delay fits both its retry ceiling and the remaining overall sidecar deadline. A delay that cannot fit returns and records the original 429 so pool routing retains quota evidence.
+A forward OpenAI search sidecar retries a 429 only when the requested delay fits both its retry ceiling and the remaining overall sidecar deadline. A delay that cannot fit returns and records the original 429 so pool routing retains quota evidence. The classic OpenAI search helper in `src/web-search/executor.ts` removes the Responses Lite marker from merged configured and selected headers before sending its hosted-search request, while retaining other forwarded headers. `tests/web-search/web-search-sidecar-429.test.ts` covers both header sources.
 One search makes at most three physical sends in total: connection-reset recovery and 429 replays draw from the same budget, and a budget spent with a 429 in hand ends with that 429 as the recorded outcome.
 OpenAI helpers recheck the resolved [stored-account credit policy](codex-account-controls.md#stored-account-authentication-policy) immediately before every physical send, including those retries. An unchanged caller-owned Direct credential is not governed by stored-pool consent.
 An explicitly unfinished tool end is passed through and never executes. A leg whose

tests/web-search/web-search-sidecar-429.test.ts

@@ -1,7 +1,7 @@
 import { afterAll, afterEach, describe, expect, test } from "bun:test";
 import { runWebSearch as runOpenAiWebSearch } from "../../src/web-search/executor";
 import { listOpenAiForwardSidecarCandidates } from "../../src/providers/openai-sidecar";
-import type { OcxConfig } from "../../src/types";
+import type { OcxConfig, OcxProviderConfig } from "../../src/types";
 
 const originalFetch = globalThis.fetch;
 afterEach(() => { globalThis.fetch = originalFetch; });
@@ -174,3 +174,69 @@ describe("web-search sidecar 429 replays", () => {
     expect(recorded).toEqual(["connect_neutral"]);
   });
 });
+
+test("issue #8160 - OpenAI search helper strips responses-lite marker from merged headers", async () => {
+  const originalFetch = globalThis.fetch;
+  let capturedHeaders: Headers | undefined;
+  globalThis.fetch = (async (_url: any, init: any) => {
+    capturedHeaders = new Headers(init?.headers);
+    return new Response(
+      'data: ' + JSON.stringify({ type: "response.completed", response: { status: "completed", output: [] } }) + '\n\n',
+      { headers: { "content-type": "text/event-stream" } },
+    );
+  }) as typeof fetch;
+
+  try {
+    await runOpenAiWebSearch(
+      "synthetic query",
+      { type: "web_search" },
+      {
+        adapter: "openai-responses",
+        authMode: "forward",
+        baseUrl: "https://chatgpt.com/backend-api/codex",
+        headers: { "x-openai-internal-codex-responses-lite": "true", "x-custom-test": "forwarded" },
+      } as any,
+      new Headers({
+        "x-openai-internal-codex-responses-lite": "true",
+        "chatgpt-account-id": "test-account",
+      }),
+      { model: "gpt-6.1-sol", reasoning: "low", timeoutMs: 1_000 },
+    );
+
+    expect(capturedHeaders).toBeDefined();
+    expect(capturedHeaders!.has("x-openai-internal-codex-responses-lite")).toBe(false);
+    expect(capturedHeaders!.get("chatgpt-account-id")).toBe("test-account");
+    expect(capturedHeaders!.get("x-custom-test")).toBe("forwarded");
+  } finally {
+    globalThis.fetch = originalFetch;
+  }
+});
+
+test.each([
+  { name: "configured header only", configured: true, selected: false },
+  { name: "selected header only", configured: false, selected: true },
+])("classic search strips the Responses Lite marker from $name", async ({ configured, selected }) => {
+  let sent: Headers | undefined;
+  globalThis.fetch = (async (_url: unknown, init?: RequestInit) => {
+    sent = new Headers(init?.headers);
+    return new Response('data: {"type":"response.completed","response":{"status":"completed","output":[]}}\n\n');
+  }) as typeof fetch;
+
+  await runOpenAiWebSearch(
+    "synthetic query",
+    { type: "web_search" },
+    {
+      adapter: "openai-responses",
+      authMode: "forward",
+      baseUrl: "https://chatgpt.com/backend-api/codex",
+      headers: { ...(configured ? { "X-OpenAI-Internal-Codex-Responses-Lite": "true" } : {}), "x-custom-test": "forwarded" },
+    } as OcxProviderConfig,
+    new Headers({ ...(selected ? { "X-OpenAI-Internal-Codex-Responses-Lite": "true" } : {}), "chatgpt-account-id": "test-account" }),
+    { model: "gpt-6.1-sol", reasoning: "low", timeoutMs: 1_000 },
+  );
+
+  expect(sent).toBeDefined();
+  expect(sent!.has("x-openai-internal-codex-responses-lite")).toBe(false);
+  expect(sent!.get("chatgpt-account-id")).toBe("test-account");
+  expect(sent!.get("x-custom-test")).toBe("forwarded");
+});