diff --git a/scripts/e2e/openai-chat-tools-docker.sh b/scripts/e2e/openai-chat-tools-docker.sh
index 57039abc2bff..00fdfcb61100 100644
--- a/scripts/e2e/openai-chat-tools-docker.sh
+++ b/scripts/e2e/openai-chat-tools-docker.sh
@@ -8,6 +8,9 @@ IMAGE_NAME="$(docker_e2e_resolve_image "openclaw-openai-chat-tools-e2e" OPENCLAW
 SKIP_BUILD="${OPENCLAW_OPENAI_CHAT_TOOLS_E2E_SKIP_BUILD:-0}"
 PORT="$(docker_e2e_read_tcp_port_env OPENCLAW_OPENAI_CHAT_TOOLS_PORT 18789)"
 TIMEOUT_SECONDS="$(docker_e2e_read_positive_int_env OPENCLAW_OPENAI_CHAT_TOOLS_TIMEOUT_SECONDS 180)"
+MAX_BODY_BYTES="$(
+  docker_e2e_read_positive_int_env OPENCLAW_OPENAI_CHAT_TOOLS_MAX_BODY_BYTES 1048576
+)"
 TOKEN="openai-chat-tools-e2e-$$"
 PROFILE_FILE="${OPENCLAW_OPENAI_CHAT_TOOLS_PROFILE_FILE:-${OPENCLAW_TESTBOX_PROFILE_FILE:-$HOME/.openclaw-testbox-live.profile}}"
 if [ ! -f "$PROFILE_FILE" ] && [ -f "$HOME/.profile" ]; then
@@ -65,6 +68,7 @@ docker_e2e_run_logged_with_harness openai-chat-tools \
   -e "OPENCLAW_GATEWAY_TOKEN=$TOKEN" \
   -e "OPENCLAW_OPENAI_CHAT_TOOLS_MODEL=${OPENCLAW_OPENAI_CHAT_TOOLS_MODEL:-openai/gpt-5.4-mini}" \
   -e "OPENCLAW_OPENAI_CHAT_TOOLS_TIMEOUT_SECONDS=$TIMEOUT_SECONDS" \
+  -e "OPENCLAW_OPENAI_CHAT_TOOLS_MAX_BODY_BYTES=$MAX_BODY_BYTES" \
   -e "OPENCLAW_TEST_STATE_SCRIPT_B64=$OPENCLAW_TEST_STATE_SCRIPT_B64" \
   -e "PORT=$PORT" \
   "${PROFILE_MOUNT[@]}" \
diff --git a/test/scripts/openai-chat-tools-client.test.ts b/test/scripts/openai-chat-tools-client.test.ts
index d270f969f90b..1f6b9020e756 100644
--- a/test/scripts/openai-chat-tools-client.test.ts
+++ b/test/scripts/openai-chat-tools-client.test.ts
@@ -6,6 +6,7 @@ import { tmpdir } from "node:os";
 import path from "node:path";
 import { beforeAll, describe, expect, it } from "vitest";
 import { createBoundedChildOutput } from "../helpers/bounded-child-output.js";
+import { cleanupTempDirs, makeTempDir } from "../helpers/temp-dir.js";
 
 const clientPath = path.resolve("scripts/e2e/lib/openai-chat-tools/client.mjs");
 const dockerRunnerPath = path.resolve("scripts/e2e/openai-chat-tools-docker.sh");
@@ -204,6 +205,44 @@ describe("scripts/e2e/lib/openai-chat-tools/client.mjs", () => {
       rmSync(root, { force: true, recursive: true });
     }
   });
+
+  it.each([
+    ["timeout", "OPENCLAW_OPENAI_CHAT_TOOLS_TIMEOUT_SECONDS", "1e3"],
+    ["body limit", "OPENCLAW_OPENAI_CHAT_TOOLS_MAX_BODY_BYTES", "64bytes"],
+  ])(
+    "rejects invalid Docker runner %s before auth or Docker build work starts",
+    (_label, envName, value) => {
+      const tempDirs: string[] = [];
+      const root = makeTempDir(tempDirs, "openclaw-openai-chat-tools-");
+      try {
+        const result = runDockerRunnerAuthPreflight(root, { [envName]: value });
+        const output = `${result.stdout}\n${result.stderr}`;
+
+        expect(result.status).toBe(2);
+        expect(output).toContain(`invalid ${envName}: ${value}`);
+        expect(output).not.toContain("OPENAI_API_KEY was not available");
+        expect(output).not.toContain("Building Docker image:");
+        expect(output).not.toContain("Reusing Docker image:");
+        expect(output).not.toContain("Running OpenAI Chat Completions tools Docker E2E");
+      } finally {
+        cleanupTempDirs(tempDirs);
+      }
+    },
+  );
+
+  it("passes normalized timeout and body limits into the Docker runner", () => {
+    const runner = readFileSync(dockerRunnerPath, "utf8");
+
+    expect(runner).toContain(
+      "docker_e2e_read_positive_int_env OPENCLAW_OPENAI_CHAT_TOOLS_TIMEOUT_SECONDS 180",
+    );
+    expect(runner).toContain(
+      "docker_e2e_read_positive_int_env OPENCLAW_OPENAI_CHAT_TOOLS_MAX_BODY_BYTES 1048576",
+    );
+    expect(runner).toContain('-e "OPENCLAW_OPENAI_CHAT_TOOLS_TIMEOUT_SECONDS=$TIMEOUT_SECONDS"');
+    expect(runner).toContain('-e "OPENCLAW_OPENAI_CHAT_TOOLS_MAX_BODY_BYTES=$MAX_BODY_BYTES"');
+  });
+
   it("rejects loose timeout env values instead of parsing numeric prefixes", async () => {
     const result = await runClient(1, {
       OPENCLAW_OPENAI_CHAT_TOOLS_TIMEOUT_SECONDS: "1e3",