diff --git a/open-sse/translator/concerns/thinkingUnified.js b/open-sse/translator/concerns/thinkingUnified.js
index 3ccd442f..5ce73b32 100644
--- a/open-sse/translator/concerns/thinkingUnified.js
+++ b/open-sse/translator/concerns/thinkingUnified.js
@@ -102,9 +102,28 @@ export function extractThinking(body) {
   return null;
 }
 
-// Capture thinking intent from a body. Alias of extractThinking, named for clarity
-// at the call-site where intent is snapshotted before format translation.
-export const captureThinking = extractThinking;
+// Capture thinking intent from a body before format translation strips it.
+// Besides the effort, records whether an OpenAI-shaped client wants the thinking
+// text itself: Claude returns it only with thinking.display "summarized", a field
+// OpenAI has no equivalent for, so the intent cannot survive translation on its own.
+export function captureThinking(body) {
+  const cfg = extractThinking(body);
+  if (!cfg || cfg.mode === "none") return cfg;
+  const display = openAIThinkingDisplay(body);
+  return display ? { ...cfg, display } : cfg;
+}
+
+function openAIThinkingDisplay(body) {
+  // Responses API: reasoning.summary is the explicit request for reasoning text.
+  if (body.reasoning && typeof body.reasoning === "object") {
+    const summary = body.reasoning.summary;
+    return typeof summary === "string" && summary && summary !== "none" ? "summarized" : undefined;
+  }
+  // Chat Completions has no summary knob. A client setting reasoning_effort is
+  // asking for reasoning, and reasoning_content is how it would receive it.
+  if (typeof body.reasoning_effort === "string") return "summarized";
+  return undefined;
+}
 
 const NATIVE_ONLY_FORMATS = new Set(["gemini-level", "gemini-budget", "claude-budget", "claude-adaptive", "kiro"]);
 
@@ -383,7 +402,8 @@ export function applyThinking(targetFormat, model, body, provider = null, intent
   const supportedLevels = getThinkingLevels(provider, cleanModel);
   // Anthropic's `display` (summarized | omitted) decides whether thinking text
   // comes back at all; keep what the client asked for instead of resetting it.
-  const display = typeof body.thinking?.display === "string" ? body.thinking.display : undefined;
+  // An OpenAI-shaped client's ask arrives via the captured intent instead.
+  const display = typeof body.thinking?.display === "string" ? body.thinking.display : intent?.display;
   stripAll(body);
   applyFormat(fmt, body, cfg, caps, supportedLevels, display);
   return body;
diff --git a/tests/translator/__snapshots__/golden-request.test.js.snap b/tests/translator/__snapshots__/golden-request.test.js.snap
index f5a5253b..db2c7378 100644
--- a/tests/translator/__snapshots__/golden-request.test.js.snap
+++ b/tests/translator/__snapshots__/golden-request.test.js.snap
@@ -119,6 +119,7 @@ exports[`GOLDEN request: OpenAI → Claude > reasoning_effort → adaptive outpu
     },
   ],
   "thinking": {
+    "display": "summarized",
     "type": "adaptive",
   },
 }
diff --git a/tests/translator/openai-thinking-display.test.js b/tests/translator/openai-thinking-display.test.js
new file mode 100644
index 00000000..7cbef379
--- /dev/null
+++ b/tests/translator/openai-thinking-display.test.js
@@ -0,0 +1,83 @@
+// OpenAI-format clients asking for reasoning get Claude's thinking text back.
+//
+// Claude only returns thinking text when the request sets thinking.display to
+// "summarized" — otherwise the redact-thinking beta is sent and every thinking
+// block comes back signature-only. That field has no OpenAI equivalent, so an
+// OpenAI-format client (opencode, DeepSeek Harness, Cherry Studio, ...) could never
+// see its reasoning: reasoning_content stayed empty however high the effort was.
+//
+// The client's intent is read from the pre-translation body:
+// - Chat Completions: setting reasoning_effort is the request for reasoning.
+// - Responses API: reasoning.summary is OpenAI's explicit ask for summaries.
+// Claude-format clients are untouched — they set display themselves.
+import { describe, it, expect } from "vitest";
+import "./registerAll.js";
+import { translateRequest } from "../../open-sse/translator/index.js";
+import { FORMATS } from "../../open-sse/translator/formats.js";
+import { selectAnthropicBeta } from "../../open-sse/providers/shared.js";
+
+const REDACT = "redact-thinking-2026-02-12";
+const MODEL = "claude-opus-5";
+
+function toClaude(sourceFormat, body) {
+  return translateRequest(sourceFormat, FORMATS.CLAUDE, MODEL, body, true, null, "claude");
+}
+
+const chat = (extra = {}) => ({ model: MODEL, messages: [{ role: "user", content: "Is 391 prime?" }], ...extra });
+const responses = (extra = {}) => ({ model: MODEL, input: [{ role: "user", content: "Is 391 prime?" }], ...extra });
+
+describe("Chat Completions clients", () => {
+  it("reasoning_effort asks Claude for summarized thinking and drops redact-thinking", () => {
+    const out = toClaude(FORMATS.OPENAI, chat({ reasoning_effort: "high" }));
+    expect(out.thinking.display).toBe("summarized");
+    expect(selectAnthropicBeta(MODEL, out)).not.toContain(REDACT);
+  });
+
+  it("reasoning_effort none leaves thinking off and keeps redact-thinking", () => {
+    const out = toClaude(FORMATS.OPENAI, chat({ reasoning_effort: "none" }));
+    expect(out.thinking?.display).toBeUndefined();
+    expect(selectAnthropicBeta(MODEL, out)).toContain(REDACT);
+  });
+
+  it("no reasoning_effort changes nothing", () => {
+    const out = toClaude(FORMATS.OPENAI, chat());
+    expect(out.thinking?.display).toBeUndefined();
+    expect(selectAnthropicBeta(MODEL, out)).toContain(REDACT);
+  });
+});
+
+describe("Responses API clients", () => {
+  it("reasoning.summary asks Claude for summarized thinking", () => {
+    const out = toClaude(FORMATS.OPENAI_RESPONSES, responses({ reasoning: { effort: "high", summary: "auto" } }));
+    expect(out.thinking.display).toBe("summarized");
+    expect(selectAnthropicBeta(MODEL, out)).not.toContain(REDACT);
+  });
+
+  it("effort without summary keeps thinking text redacted, as OpenAI would", () => {
+    const out = toClaude(FORMATS.OPENAI_RESPONSES, responses({ reasoning: { effort: "high" } }));
+    expect(out.thinking?.display).toBeUndefined();
+    expect(selectAnthropicBeta(MODEL, out)).toContain(REDACT);
+  });
+});
+
+describe("Claude-format clients are untouched", () => {
+  it("thinking without display stays redacted", () => {
+    const out = toClaude(FORMATS.CLAUDE, {
+      model: MODEL, max_tokens: 4096,
+      thinking: { type: "enabled", budget_tokens: 2048 },
+      messages: [{ role: "user", content: "Is 391 prime?" }],
+    });
+    expect(out.thinking?.display).toBeUndefined();
+    expect(selectAnthropicBeta(MODEL, out)).toContain(REDACT);
+  });
+
+  it("an explicit display is kept as sent", () => {
+    const out = toClaude(FORMATS.CLAUDE, {
+      model: MODEL, max_tokens: 4096,
+      thinking: { type: "enabled", budget_tokens: 2048, display: "omitted" },
+      messages: [{ role: "user", content: "Is 391 prime?" }],
+    });
+    expect(out.thinking.display).toBe("omitted");
+    expect(selectAnthropicBeta(MODEL, out)).toContain(REDACT);
+  });
+});