fix(openai): refresh live prompt overlay config

CHANGELOG.md

@@ -112,6 +112,7 @@ Docs: https://docs.openclaw.ai
 - Active Memory: stop reviving removed live `active-memory` config from startup snapshots, so removing the plugin entry turns the hook off immediately instead of waiting for a restart. Thanks @vincentkoc.
 - GitHub Copilot: re-read plugin discovery config from the live runtime snapshot, so toggling `plugins.entries.github-copilot.config.discovery.enabled` takes effect without a restart. Thanks @vincentkoc.
 - Ollama: re-read plugin discovery config from the live runtime snapshot, so toggling `plugins.entries.ollama.config.discovery.enabled` takes effect without a restart. Thanks @vincentkoc.
+- OpenAI: re-read the plugin prompt-overlay personality from live runtime config, so GPT-5 system prompt contributions update without a restart when `plugins.entries.openai.config.personality` changes. Thanks @vincentkoc.
 - Agents/subagents: drop bare `NO_REPLY` from the parent turn when the session still has pending spawned children, so direct-conversation surfaces such as Telegram DMs no longer rewrite the sentinel into visible fallback chatter while waiting for the child completion event. (#69942) Thanks @neeravmakwana.
 - Plugins/install: keep bundled plugin dependencies off npm install while repairing them when plugins activate from a packaged install, including Feishu/Lark, Browser, and direct bundled channel setup-entry loads.
 - CLI/channels: skip and cache bundled channel plugin, setup, and secrets load failures during read-only discovery, so one broken unused bundled channel cannot crash `openclaw status` or bootstrap secret scans.

extensions/openai/index.test.ts

@@ -563,6 +563,42 @@ describe("openai plugin", () => {
     });
   });
 
+  it("uses live plugin config for GPT-5 prompt overlay mode", async () => {
+    const { providers } = await registerOpenAIPluginWithHook({
+      pluginConfig: { personality: "off" },
+    });
+
+    const openaiProvider = requireRegisteredProvider(providers, "openai");
+    expect(
+      openaiProvider.resolveSystemPromptContribution?.({
+        config: {
+          plugins: {
+            entries: {
+              openai: {
+                config: {
+                  personality: "friendly",
+                },
+              },
+            },
+          },
+        },
+        agentDir: undefined,
+        workspaceDir: undefined,
+        provider: "openai",
+        modelId: "gpt-5.4",
+        promptMode: "full",
+        runtimeChannel: undefined,
+        runtimeCapabilities: undefined,
+        agentId: undefined,
+      }),
+    ).toEqual({
+      stablePrefix: OPENAI_GPT5_BEHAVIOR_CONTRACT,
+      sectionOverrides: {
+        interaction_style: OPENAI_FRIENDLY_PROMPT_OVERLAY,
+      },
+    });
+  });
+
   it("treats on as an alias for the friendly prompt overlay", async () => {
     const { providers } = await registerOpenAIPluginWithHook({
       pluginConfig: { personality: "on" },

extensions/openai/index.ts

@@ -1,3 +1,4 @@
+import { resolvePluginConfigObject } from "openclaw/plugin-sdk/config-runtime";
 import { definePluginEntry } from "openclaw/plugin-sdk/plugin-entry";
 import { buildProviderToolCompatFamilyHooks } from "openclaw/plugin-sdk/provider-tools";
 import { buildOpenAICodexCliBackend } from "./cli-backend.js";
@@ -23,21 +24,25 @@ export default definePluginEntry({
   name: "OpenAI Provider",
   description: "Bundled OpenAI provider plugins",
   register(api) {
-    const promptOverlayMode = resolveOpenAIPromptOverlayMode(api.pluginConfig);
     const openAIToolCompatHooks = buildProviderToolCompatFamilyHooks("openai");
     const buildProviderWithPromptContribution = <T extends ReturnType<typeof buildOpenAIProvider>>(
       provider: T,
     ): T => ({
       ...provider,
       ...openAIToolCompatHooks,
-      resolveSystemPromptContribution: (ctx) =>
-        resolveOpenAISystemPromptContribution({
+      resolveSystemPromptContribution: (ctx) => {
+        const runtimePluginConfig = resolvePluginConfigObject(ctx.config, "openai");
+        const pluginConfig =
+          runtimePluginConfig ??
+          (ctx.config ? undefined : (api.pluginConfig as Record<string, unknown>));
+        return resolveOpenAISystemPromptContribution({
           config: ctx.config,
-          legacyPluginConfig: api.pluginConfig,
-          mode: promptOverlayMode,
+          legacyPluginConfig: pluginConfig,
+          mode: resolveOpenAIPromptOverlayMode(pluginConfig),
           modelProviderId: provider.id,
           modelId: ctx.modelId,
-        }),
+        });
+      },
     });
     api.registerCliBackend(buildOpenAICodexCliBackend());
     api.registerProvider(buildProviderWithPromptContribution(buildOpenAIProvider()));