feat: add provider-native request switches (#5254)

docs/configuration.md

@@ -347,6 +347,36 @@ Valid `apiType` values are exactly `auto`, `chat_completions`, and `responses`.
 }
 ```
 
+The WebUI's OpenAI web-search switch writes the corresponding `apiType` and `extraBody.tools`
+fields. A hosted search tool replaces nanobot's same-name local `web_search` function for that
+request, while other tools such as `web_fetch` remain available.
+
+</details>
+
+<details>
+<summary><b>DeepSeek native web search</b></summary>
+
+DeepSeek V4 Flash uses DeepSeek's native Responses API. Its provider-hosted web search is
+enabled by default because it does not require a separate paid add-on. Turn it off from the
+WebUI provider settings, or with:
+
+```json
+{
+  "providers": {
+    "deepseek": {
+      "apiKey": "${DEEPSEEK_API_KEY}",
+      "extraBody": {
+        "tools": []
+      }
+    }
+  }
+}
+```
+
+The switch applies to `deepseek-v4-flash`; DeepSeek models that remain on Chat Completions
+cannot use this Responses tool. Native search calls appear in the WebUI activity stream, and
+their opaque output items are preserved for multi-turn Responses state replay.
+
 </details>
 
 <a id="responses-state-and-compaction"></a>
@@ -695,7 +725,7 @@ Then run:
 nanobot agent -m "Hello!"
 ```
 
-To opt in to Codex Fast mode, merge this provider setting into `config.json`:
+Codex Fast mode can be enabled from the WebUI provider settings, or with:
 
 ```json
 {
@@ -709,9 +739,9 @@ To opt in to Codex Fast mode, merge this provider setting into `config.json`:
 }
 ```
 
-`priority` is the Responses API request value used by Codex Fast mode. The setting only works
+The switch sends the Responses API `service_tier: "priority"` value. It only works for models
-for models and accounts that support Fast mode; remove `service_tier` to return to standard
+and accounts that support Fast mode; turn the switch off to return to standard processing.
-processing. Fast mode consumes Codex credits at a higher rate. See the
+Fast mode consumes Codex credits at a higher rate. See the
 [OpenAI Codex rate card](https://help.openai.com/en/articles/20001106) for current details.
 
 For proxy, remote/headless login, model-name, or config-key errors, see [`troubleshooting.md`](./troubleshooting.md#provider-and-model-problems).
@@ -735,6 +765,8 @@ The provider reads xAI's model catalog and includes the server-hosted `x_search`
 tool only when the selected model advertises `supportsBackendSearch`. Models
 without that capability continue normally without hosted X Search. When enabled,
 searches run inside xAI's Responses API and citations arrive as inline links.
+Hosted X Search is on by default to preserve this behavior. It can be turned off in the
+WebUI provider settings or with `providers.xaiGrok.extraBody.tools: []`.
 
 This is xAI subscription OAuth, not X Developer OAuth. nanobot follows the
 public OAuth client and proxy contract used by

docs/providers.md

@@ -262,9 +262,9 @@ Arbitrary custom provider names are OpenAI-compatible only; they do not use the
 }
 ```
 
-`providers.openai.apiType` may be set when you need to force a specific OpenAI API surface. Other providers reject `apiType`; leave it unset outside `providers.openai`. Replace the model with a model ID available to your OpenAI account. Direct OpenAI Responses, OpenAI Codex, Azure OpenAI Responses, and eligible GitHub Copilot models share [opaque Responses state retention](./configuration.md#responses-state-and-compaction); native compaction is enabled only where the backend supports it.
+`providers.openai.apiType` may be set when you need to force a specific OpenAI API surface. Other providers reject `apiType`; leave it unset outside `providers.openai`. Replace the model with a model ID available to your OpenAI account. Direct OpenAI Responses, OpenAI Codex, Azure OpenAI Responses, and eligible GitHub Copilot models share [opaque Responses state retention](./configuration.md#responses-state-and-compaction); native compaction is enabled only where the backend supports it. The WebUI exposes provider-native switches for OpenAI web search, Codex Fast mode, DeepSeek web search, and Grok X Search. These switches write the corresponding raw provider request fields under `extraBody`.
 
-DeepSeek is the model-level exception in the OpenAI-compatible provider: `deepseek-v4-flash` automatically uses DeepSeek's native Responses API, while `deepseek-v4-pro` remains on Chat Completions.
+DeepSeek is the model-level exception in the OpenAI-compatible provider: `deepseek-v4-flash` automatically uses DeepSeek's native Responses API, while `deepseek-v4-pro` remains on Chat Completions. Its native `web_search` tool is enabled by default and shows its lifecycle in WebUI chat activity; set `providers.deepseek.extraBody.tools` to `[]` to disable it.
 
 ### Custom OpenAI-Compatible Endpoint
 
@@ -528,6 +528,8 @@ When enabled, Grok can search current X posts and return inline source links
 without invoking a local nanobot tool. Credentials are stored under the
 active instance's `auth/xai.json` (normally `~/.nanobot/auth/xai.json`), not in
 `config.json` and not in Grok Build's credential file.
+Hosted X Search remains enabled by default and can be disabled with the WebUI
+switch or `providers.xaiGrok.extraBody.tools: []`.
 
 The login is xAI subscription OAuth, not X Developer OAuth. It follows the
 public client contract documented and implemented by

nanobot/providers/openai_compat_provider.py

@@ -56,6 +56,32 @@ if TYPE_CHECKING:
 # that ``unittest.mock.patch`` can find and replace it.
 AsyncOpenAI: Any = None
 
+
+def _is_hosted_web_search_type(value: object) -> bool:
+    return isinstance(value, str) and (
+        value == "web_search" or value.startswith("web_search_")
+    )
+
+
+def _is_hosted_web_search_tool(tool: object) -> bool:
+    if not isinstance(tool, dict):
+        return False
+    tool_type = cast(dict[object, object], tool).get("type")
+    return _is_hosted_web_search_type(tool_type)
+
+
+def _is_named_function_tool(tool: object, name: str) -> bool:
+    """Return whether a Responses tool is a function with the given name."""
+    if not isinstance(tool, dict):
+        return False
+    record = cast(dict[object, object], tool)
+    if record.get("type") != "function":
+        return False
+    function = record.get("function")
+    if isinstance(function, dict):
+        return cast(dict[object, object], function).get("name") == name
+    return record.get("name") == name
+
 _ALLOWED_MSG_KEYS = frozenset({
     "role", "content", "tool_calls", "tool_call_id", "name",
     "reasoning_content", "extra_content",
@@ -469,7 +495,7 @@ class OpenAICompatProvider(LLMProvider):
         self.default_model = default_model
         self.extra_headers = extra_headers or {}
         self._spec = spec
-        self._extra_body = extra_body or {}
+        self._extra_body = dict(extra_body or {})
         self._api_type = api_type if spec and spec.name == "openai" else "auto"
         self._extra_query = extra_query or {}
         self._proxy = proxy or None
@@ -974,8 +1000,8 @@ class OpenAICompatProvider(LLMProvider):
         provider_responses = spec_name in ("openai", "github_copilot")
         if not provider_responses and not model_responses:
             return False
-        if self._api_type == "responses":
+        if self._responses_is_required():
-            # Explicit configuration means Responses is mandatory; do not
+            # Explicit Responses-only request fields are mandatory; do not
             # consult the circuit breaker or fall back to Chat Completions.
             return True
         if provider_responses and (self._spec is None or self._spec.name != "github_copilot"):
@@ -994,6 +1020,25 @@ class OpenAICompatProvider(LLMProvider):
 
         return self._responses_circuit_allows_probe(model, reasoning_effort)
 
+    def _responses_is_required(self) -> bool:
+        return self._api_type == "responses" or self._hosted_web_search_enabled()
+
+    def _hosted_web_search_enabled(self) -> bool:
+        extra_body = getattr(self, "_extra_body", {})
+        configured_tools = extra_body.get("tools")
+        if "tools" in extra_body:
+            return isinstance(configured_tools, list) and any(
+                _is_hosted_web_search_tool(tool)
+                for tool in cast(list[object], configured_tools)
+            )
+        return bool(
+            self._spec
+            and any(
+                _is_hosted_web_search_type(tool_type)
+                for tool_type in getattr(self._spec, "responses_default_tools", ())
+            )
+        )
+
     def _responses_state_provider(self) -> str:
         spec_name = self._spec.name if self._spec is not None else "custom"
         effective_base = self._effective_base or "https://api.openai.com/v1"
@@ -1157,9 +1202,38 @@ class OpenAICompatProvider(LLMProvider):
             body["tool_choice"] = tool_choice or "auto"
 
         extra_body = getattr(self, "_extra_body", {})
+        default_tools = getattr(self._spec, "responses_default_tools", ())
+        if "tools" not in extra_body and default_tools:
+            body["tools"] = [
+                *cast(list[object], body.get("tools", [])),
+                *({"type": tool_type} for tool_type in default_tools),
+            ]
         if extra_body:
             body = _merge_responses_extra_body(body, extra_body)
 
+        if self._hosted_web_search_enabled():
+            configured_tools = body.get("tools")
+            if isinstance(configured_tools, list):
+                managed_tools: list[object] = []
+                hosted_search_seen = False
+                for tool in cast(list[object], configured_tools):
+                    if _is_named_function_tool(tool, "web_search"):
+                        continue
+                    if _is_hosted_web_search_tool(tool):
+                        if hosted_search_seen:
+                            continue
+                        hosted_search_seen = True
+                    managed_tools.append(tool)
+                body["tools"] = managed_tools
+            if self._spec and self._spec.name == "openai":
+                source_include = "web_search_call.action.sources"
+                configured_include = body.get("include")
+                if isinstance(configured_include, list):
+                    if source_include not in configured_include:
+                        body["include"] = [*configured_include, source_include]
+                else:
+                    body["include"] = [source_include]
+
         return body
 
     async def _create_response_with_compaction_fallback(
@@ -1771,7 +1845,7 @@ class OpenAICompatProvider(LLMProvider):
                         # falling back to /chat/completions cannot succeed and would
                         # hide the real error.
                         raise
-                    if self._api_type == "responses":
+                    if self._responses_is_required():
                         raise
                     if not self._should_fallback_from_responses_error(responses_error):
                         raise
@@ -1867,7 +1941,7 @@ class OpenAICompatProvider(LLMProvider):
                         # falling back to /chat/completions cannot succeed and would
                         # hide the real error.
                         raise
-                    if self._api_type == "responses":
+                    if self._responses_is_required():
                         raise
                     if not self._should_fallback_from_responses_error(responses_error):
                         raise

nanobot/providers/openai_responses/parsing.py

@@ -89,6 +89,77 @@ def _response_object_list(value: object) -> list[dict[str, Any]]:
     ]
 
 
+def _hosted_web_search_event(
+    event: object,
+    event_type: object,
+) -> dict[str, Any] | None:
+    """Map the official web-search output item pair onto normal tool progress."""
+    if event_type not in {"response.output_item.added", "response.output_item.done"}:
+        return None
+    event_object = _response_object(event) or {}
+    item = _response_object(event_object.get("item")) or {}
+    if item.get("type") != "web_search_call":
+        return None
+    call_id = item.get("id") or item.get("call_id") or event_object.get("item_id")
+    if not isinstance(call_id, str) or not call_id:
+        return None
+
+    action = _response_object(item.get("action")) or {}
+    raw_queries = action.get("queries")
+    queries = (
+        [
+            query.strip()
+            for query in cast(list[object], raw_queries)
+            if isinstance(query, str) and query.strip()
+        ][:4]
+        if isinstance(raw_queries, list)
+        else []
+    )
+    query = " · ".join(queries)
+    if not query:
+        query = next(
+            (
+                value.strip()
+                for key in ("query", "pattern", "url")
+                if isinstance((value := action.get(key)), str) and value.strip()
+            ),
+            "",
+        )
+    arguments = {"query": query[:1000]} if query else {}
+
+    phase = "start" if event_type == "response.output_item.added" else "end"
+    result: dict[str, Any] | None = None
+    if phase == "end":
+        status = item.get("status")
+        result = {"status": status if isinstance(status, str) else "completed"}
+        raw_sources = action.get("sources")
+        if isinstance(raw_sources, list):
+            sources: list[dict[str, str]] = []
+            for raw_source in cast(list[object], raw_sources):
+                source = _response_object(raw_source) or {}
+                url = source.get("url")
+                if not isinstance(url, str) or not url.strip():
+                    continue
+                visible_source = {"url": url.strip()[:2048]}
+                title = source.get("title")
+                if isinstance(title, str) and title.strip():
+                    visible_source["title"] = title.strip()[:300]
+                sources.append(visible_source)
+                if len(sources) == 8:
+                    break
+            if sources:
+                result["sources"] = sources
+
+    return {
+        "kind": "hosted_tool",
+        "phase": phase,
+        "call_id": call_id,
+        "name": "web_search",
+        "arguments": arguments,
+        "result": result,
+    }
+
+
 def map_finish_reason(status: str | None) -> str:
     """Map a Responses API status string to a Chat-Completions-style finish_reason."""
     return FINISH_REASON_MAP.get(status or "completed", "stop")
@@ -269,11 +340,14 @@ async def consume_sse_with_reasoning(
     refusal_seen = False
     refusal_deltas: dict[tuple[str | None, int | None], str] = {}
     emitted_refusal_text = ""
-
     async for event in iter_sse(response):
         if on_response_event:
             await on_response_event(event)
         event_type = event.get("type")
+        if on_tool_call_delta and (
+            hosted_event := _hosted_web_search_event(event, event_type)
+        ):
+            await on_tool_call_delta(hosted_event)
         if event_type == "response.output_item.added":
             item = _as_json_object(event.get("item")) or {}
             if item.get("type") == "function_call":
@@ -555,10 +629,13 @@ async def consume_sdk_stream(
     refusal_seen = False
     refusal_deltas: dict[tuple[str | None, int | None], str] = {}
     emitted_refusal_text = ""
-
     async for raw_event in stream:
         event: Any = raw_event
         event_type = getattr(event, "type", None)
+        if on_tool_call_delta and (
+            hosted_event := _hosted_web_search_event(event, event_type)
+        ):
+            await on_tool_call_delta(hosted_event)
         if event_type == "response.output_item.added":
             item = getattr(event, "item", None)
             if item and getattr(item, "type", None) == "function_call":

nanobot/providers/registry.py

@@ -116,6 +116,10 @@ class ProviderSpec:
     # Flash is supported before V4 Pro).
     responses_models: tuple[str, ...] = ()
 
+    # Provider-hosted Responses tools sent unless extraBody.tools explicitly
+    # supplies the hosted-tool selection. Values are raw Responses tool types.
+    responses_default_tools: tuple[str, ...] = ()
+
     # When the model returns content as a list of {"type":"thinking",...} +
     # {"type":"text",...} blocks, extract the thinking text into
     # reasoning_content. Mistral's Magistral / reasoning-enabled responses use
@@ -479,6 +483,7 @@ PROVIDERS: tuple[ProviderSpec, ...] = (
         default_api_base="https://api.deepseek.com",
         thinking_style="thinking_type",
         responses_models=("deepseek-v4-flash",),
+        responses_default_tools=("web_search",),
     ),
     # Gemini: Google's OpenAI-compatible endpoint
     ProviderSpec(

nanobot/providers/xai_grok_provider.py

@@ -46,6 +46,19 @@ _SENSITIVE_ERROR_KEYS = {
 }
 
 
+def _is_hosted_x_search_tool(value: object) -> bool:
+    if not isinstance(value, dict):
+        return False
+    return cast(dict[object, object], value).get("type") == "x_search"
+
+
+def _is_named_x_search_tool(value: object) -> bool:
+    if not isinstance(value, dict):
+        return False
+    record = cast(dict[object, object], value)
+    return record.get("type") == "function" and record.get("name") == "x_search"
+
+
 class XAIGrokProvider(LLMProvider):
     """Call xAI's subscription proxy and expose supported hosted tools."""
 
@@ -112,13 +125,27 @@ class XAIGrokProvider(LLMProvider):
         stage = "oauth_token"
         try:
             token = await asyncio.to_thread(get_xai_oauth_token, proxy=self.proxy)
-            stage = "model_capabilities"
+            configured_tools = self._extra_body.get("tools")
-            supports_backend_search = await self._supports_backend_search(token, wire_model)
+            tools_are_explicit = "tools" in self._extra_body
+            configured_hosted_search = (
+                isinstance(configured_tools, list)
+                and any(
+                    _is_hosted_x_search_tool(tool)
+                    for tool in cast(list[object], configured_tools)
+                )
+            )
+            supports_backend_search = False
+            if not tools_are_explicit:
+                stage = "model_capabilities"
+                supports_backend_search = await self._supports_backend_search(token, wire_model)
             converted_tools = convert_tools(tools or [])
-            if supports_backend_search:
+            if isinstance(configured_tools, list):
+                converted_tools.extend(cast(list[dict[str, Any]], configured_tools))
+            if supports_backend_search or configured_hosted_search:
                 converted_tools = [
-                    tool for tool in converted_tools if tool.get("name") != "x_search"
+                    tool for tool in converted_tools if not _is_named_x_search_tool(tool)
                 ]
+            if supports_backend_search:
                 converted_tools.append({"type": "x_search"})
 
             body: dict[str, Any] = {
@@ -137,7 +164,13 @@ class XAIGrokProvider(LLMProvider):
                 "reasoning": _build_reasoning_options(reasoning_effort),
             }
             if self._extra_body:
-                body.update(self._extra_body)
+                body.update({
+                    key: value
+                    for key, value in self._extra_body.items()
+                    if key != "tools"
+                })
+                if tools_are_explicit and not isinstance(configured_tools, list):
+                    body["tools"] = configured_tools
 
             headers = _build_headers(token.access, wire_model)
             stage = "xai_request"

tests/providers/test_extra_body_config.py

@@ -241,6 +241,112 @@ class TestBuildResponsesBodyExtraBody:
             {"type": "web_search"},
         ]
 
+    def test_responses_web_search_tool_owns_the_local_function(self) -> None:
+        provider = OpenAICompatProvider(
+            api_key="test-key",
+            default_model="gpt-4o",
+            spec=find_by_name("openai"),
+            extra_body={"tools": [{"type": "web_search"}]},
+        )
+
+        body = provider._build_responses_body(
+            messages=_simple_messages(),
+            tools=[
+                {
+                    "type": "function",
+                    "function": {
+                        "name": "web_search",
+                        "description": "Search with nanobot's configured backend",
+                        "parameters": {"type": "object"},
+                    },
+                },
+                {
+                    "type": "function",
+                    "function": {
+                        "name": "read_file",
+                        "description": "Read a file",
+                        "parameters": {"type": "object"},
+                    },
+                },
+            ],
+            model=None,
+            max_tokens=100,
+            temperature=0.1,
+            reasoning_effort=None,
+            tool_choice=None,
+        )
+
+        assert body["tools"] == [
+            {
+                "type": "function",
+                "name": "read_file",
+                "description": "Read a file",
+                "parameters": {"type": "object"},
+            },
+            {"type": "web_search"},
+        ]
+        assert body["include"] == ["web_search_call.action.sources"]
+        assert provider._should_use_responses_api(None, None) is True
+
+    def test_deepseek_default_search_replaces_the_local_search_function(self) -> None:
+        provider = OpenAICompatProvider(
+            api_key="test-key",
+            default_model="deepseek-v4-flash",
+            spec=find_by_name("deepseek"),
+        )
+
+        body = provider._build_responses_body(
+            messages=_simple_messages(),
+            tools=[{
+                "type": "function",
+                "function": {
+                    "name": "web_search",
+                    "description": "Search with nanobot's configured backend",
+                    "parameters": {"type": "object"},
+                },
+            }],
+            model=None,
+            max_tokens=100,
+            temperature=0.1,
+            reasoning_effort=None,
+            tool_choice=None,
+        )
+
+        assert body["tools"] == [{"type": "web_search"}]
+        assert "include" not in body
+
+    def test_explicit_empty_tools_disables_deepseek_default_search(self) -> None:
+        provider = OpenAICompatProvider(
+            api_key="test-key",
+            default_model="deepseek-v4-flash",
+            spec=find_by_name("deepseek"),
+            extra_body={"tools": []},
+        )
+
+        body = provider._build_responses_body(
+            messages=_simple_messages(),
+            tools=[{
+                "type": "function",
+                "function": {
+                    "name": "web_search",
+                    "description": "Search with nanobot's configured backend",
+                    "parameters": {"type": "object"},
+                },
+            }],
+            model=None,
+            max_tokens=100,
+            temperature=0.1,
+            reasoning_effort=None,
+            tool_choice=None,
+        )
+
+        assert body["tools"] == [{
+            "type": "function",
+            "name": "web_search",
+            "description": "Search with nanobot's configured backend",
+            "parameters": {"type": "object"},
+        }]
+
     def test_responses_extra_body_merges_include_without_duplicates(self) -> None:
         provider = OpenAICompatProvider(
             api_key="test-key",

tests/providers/test_openai_responses.py

@@ -2,6 +2,7 @@
 
 import json
 from io import StringIO
+from types import SimpleNamespace
 from unittest.mock import MagicMock, patch
 
 import pytest
@@ -1387,6 +1388,91 @@ class TestConsumeSdkStream:
         assert tool_calls == []
         assert finish_reason == "stop"
 
+    @pytest.mark.asyncio
+    async def test_hosted_web_search_lifecycle_is_streamed_as_tool_progress(self):
+        search_added = SimpleNamespace(
+            type="web_search_call",
+            id="ws_1",
+            status="in_progress",
+            action=SimpleNamespace(type="search"),
+        )
+        search_done = SimpleNamespace(
+            type="web_search_call",
+            id="ws_1",
+            status="completed",
+            action=SimpleNamespace(
+                type="search",
+                queries=["nanobot DeepSeek", "nanobot latest release"],
+                sources=[
+                    SimpleNamespace(
+                        title="DeepSeek Responses API",
+                        url="https://api-docs.deepseek.com/guides/responses_api/",
+                    ),
+                ],
+            ),
+        )
+        response = SimpleNamespace(status="completed", usage=None, output=[search_done])
+        events = [
+            SimpleNamespace(
+                type="response.output_item.added",
+                output_index=0,
+                item=search_added,
+            ),
+            SimpleNamespace(
+                type="response.web_search_call.searching",
+                item_id="ws_1",
+                output_index=0,
+            ),
+            SimpleNamespace(
+                type="response.web_search_call.completed",
+                item_id="ws_1",
+                output_index=0,
+            ),
+            SimpleNamespace(
+                type="response.output_item.done",
+                output_index=0,
+                item=search_done,
+            ),
+            SimpleNamespace(type="response.completed", response=response),
+        ]
+        tool_events: list[dict] = []
+
+        async def stream():
+            for event in events:
+                yield event
+
+        async def on_tool_event(event: dict) -> None:
+            tool_events.append(event)
+
+        await consume_sdk_stream(stream(), on_tool_call_delta=on_tool_event)
+
+        assert tool_events == [
+            {
+                "kind": "hosted_tool",
+                "phase": "start",
+                "call_id": "ws_1",
+                "name": "web_search",
+                "arguments": {},
+                "result": None,
+            },
+            {
+                "kind": "hosted_tool",
+                "phase": "end",
+                "call_id": "ws_1",
+                "name": "web_search",
+                "arguments": {
+                    "query": "nanobot DeepSeek · nanobot latest release",
+                },
+                "result": {
+                    "status": "completed",
+                    "sources": [{
+                        "title": "DeepSeek Responses API",
+                        "url": "https://api-docs.deepseek.com/guides/responses_api/",
+                    }],
+                },
+            },
+        ]
+
     @pytest.mark.asyncio
     async def test_refusal_events_reconcile_parts_and_terminal_output(self):
         refusal = "First and second sentence. Done-only. Terminal suffix."

tests/providers/test_xai_grok_provider.py

@@ -139,6 +139,111 @@ async def test_provider_injects_hosted_x_search_and_required_proxy_headers(monke
     assert headers["x-grok-model-override"] == "grok-4.5"
 
 
+@pytest.mark.asyncio
+async def test_explicit_parameterized_x_search_is_preserved_without_catalog_lookup(
+    monkeypatch,
+) -> None:
+    _mock_token(monkeypatch)
+    bodies: list[dict[str, Any]] = []
+
+    async def unexpected_catalog_lookup(*_args, **_kwargs):
+        raise AssertionError("explicit raw tools must not depend on model catalog metadata")
+
+    async def fake_request(_url, _headers, body, **_kwargs):
+        bodies.append(body)
+        return "ok", [], "stop", {}, None
+
+    monkeypatch.setattr(
+        "nanobot.providers.xai_grok_provider._fetch_xai_model_capabilities",
+        unexpected_catalog_lookup,
+    )
+    monkeypatch.setattr("nanobot.providers.xai_grok_provider._request_xai", fake_request)
+    hosted_tool = {
+        "type": "x_search",
+        "allowed_x_handles": ["nanobot_ai"],
+        "from_date": "2026-01-01",
+    }
+    provider = XAIGrokProvider(extra_body={
+        "parallel_tool_calls": False,
+        "tools": [hosted_tool, {"type": "code_interpreter", "container": "auto"}],
+    })
+
+    response = await provider.chat(
+        [{"role": "user", "content": "search"}],
+        tools=[
+            {
+                "type": "function",
+                "function": {
+                    "name": "read_file",
+                    "description": "Read a file",
+                    "parameters": {"type": "object"},
+                },
+            },
+            {
+                "type": "function",
+                "function": {
+                    "name": "x_search",
+                    "description": "A colliding local tool",
+                    "parameters": {"type": "object"},
+                },
+            },
+        ],
+    )
+
+    assert response.content == "ok"
+    assert bodies[0]["parallel_tool_calls"] is False
+    assert bodies[0]["tools"] == [
+        {
+            "type": "function",
+            "name": "read_file",
+            "description": "Read a file",
+            "parameters": {"type": "object"},
+        },
+        hosted_tool,
+        {"type": "code_interpreter", "container": "auto"},
+    ]
+
+
+@pytest.mark.asyncio
+async def test_explicit_empty_tools_disables_catalog_lookup_and_hosted_tool(monkeypatch) -> None:
+    _mock_token(monkeypatch)
+    bodies: list[dict[str, Any]] = []
+
+    async def unexpected_catalog_lookup(*_args, **_kwargs):
+        raise AssertionError("explicitly disabled X Search must not fetch model capabilities")
+
+    async def fake_request(_url, _headers, body, **_kwargs):
+        bodies.append(body)
+        return "ok", [], "stop", {}, None
+
+    monkeypatch.setattr(
+        "nanobot.providers.xai_grok_provider._fetch_xai_model_capabilities",
+        unexpected_catalog_lookup,
+    )
+    monkeypatch.setattr("nanobot.providers.xai_grok_provider._request_xai", fake_request)
+    provider = XAIGrokProvider(extra_body={"tools": []})
+
+    response = await provider.chat(
+        [{"role": "user", "content": "hello"}],
+        tools=[{
+            "type": "function",
+            "function": {
+                "name": "read_file",
+                "description": "Read a file",
+                "parameters": {"type": "object"},
+            },
+        }],
+    )
+
+    assert response.content == "ok"
+    assert bodies[0]["tools"] == [{
+        "type": "function",
+        "name": "read_file",
+        "description": "Read a file",
+        "parameters": {"type": "object"},
+    }]
+
+
 @pytest.mark.asyncio
 async def test_provider_keeps_local_x_search_when_model_does_not_support_hosted_search(
     monkeypatch,

tests/webui/test_settings_api.py

@@ -733,15 +733,19 @@ def test_update_provider_settings_updates_and_clears_oauth_proxy(
         },
     )
 
-    payload = update_provider_settings(
+    payload = update_provider_settings({
-        {"provider": [provider_name], "proxy": [" http://127.0.0.1:7890 "]}
+        "provider": [provider_name],
-    )
+        "proxy": [" http://127.0.0.1:7890 "],
+        "extraBody": [json.dumps({"tools": []})],
+    })
 
     providers = {row["name"]: row for row in payload["providers"]}
     assert providers[provider_name]["proxy"] == "http://127.0.0.1:7890"
     assert getattr(load_config(config_path).providers, config_attr).proxy == (
         "http://127.0.0.1:7890"
     )
+    assert providers[provider_name]["extra_body"] == {"tools": []}
+    assert getattr(load_config(config_path).providers, config_attr).extra_body == {"tools": []}
 
     cleared = update_provider_settings({"provider": [provider_name], "proxy": ["  "]})
 

webui/src/components/settings/SettingsView.tsx

@@ -276,6 +276,52 @@ type CustomMcpTransport = "stdio" | "streamableHttp" | "sse";
 
 const CONTEXT_WINDOW_TOKEN_OPTIONS = [65_536, 200_000, 262_144, 500_000, 1_048_576] as const;
 const OAUTH_PROXY_PROVIDERS = new Set(["openai_codex", "xai_grok"]);
+type ProviderRequestOption = {
+  kind: "priority" | "hosted_tool";
+  titleKey: string;
+  title: string;
+  helpKey: string;
+  help: string;
+  toolType?: "web_search" | "x_search";
+  defaultEnabled?: boolean;
+  forceResponses?: boolean;
+};
+const PROVIDER_REQUEST_OPTIONS: Partial<Record<string, ProviderRequestOption[]>> = {
+  openai_codex: [{
+    kind: "priority",
+    titleKey: "settings.providers.capabilityFastMode",
+    title: "Fast mode",
+    helpKey: "settings.providers.capabilityFastModeHelp",
+    help: "Use OpenAI's priority service tier for faster responses. This consumes credits faster.",
+  }],
+  openai: [{
+    kind: "hosted_tool",
+    titleKey: "settings.providers.capabilityOpenAISearch",
+    title: "OpenAI web search",
+    helpKey: "settings.providers.capabilityOpenAISearchHelp",
+    help: "Allow compatible Responses API models to search the web. Search activity appears in chat.",
+    toolType: "web_search",
+    forceResponses: true,
+  }],
+  deepseek: [{
+    kind: "hosted_tool",
+    titleKey: "settings.providers.capabilityDeepSeekSearch",
+    title: "DeepSeek web search",
+    helpKey: "settings.providers.capabilityDeepSeekSearchHelp",
+    help: "Let DeepSeek V4 Flash search the web through its Responses API. Search activity appears in chat.",
+    toolType: "web_search",
+    defaultEnabled: true,
+  }],
+  xai_grok: [{
+    kind: "hosted_tool",
+    titleKey: "settings.providers.capabilityXSearch",
+    title: "X Search",
+    helpKey: "settings.providers.capabilityXSearchHelp",
+    help: "Allow supported Grok models to use xAI-hosted X Search. Search activity appears in chat.",
+    toolType: "x_search",
+    defaultEnabled: true,
+  }],
+};
 const CUSTOM_PROVIDER_CREATION_KEY = "__custom_provider__";
 const CUSTOM_PROVIDER_ADVANCED_FIELDS: ProviderAdvancedField[] = [
   "extra_headers",
@@ -306,6 +352,70 @@ function providerJsonValue(value: Record<string, unknown> | null | undefined): s
   return value && Object.keys(value).length > 0 ? JSON.stringify(value, null, 2) : "";
 }
 
+function parseProviderExtraBody(value: string): Record<string, unknown> | null {
+  if (!value.trim()) return {};
+  try {
+    const parsed: unknown = JSON.parse(value);
+    return parsed && typeof parsed === "object" && !Array.isArray(parsed)
+      ? parsed as Record<string, unknown>
+      : null;
+  } catch {
+    return null;
+  }
+}
+
+function isHostedSearchTool(tool: unknown, toolType: "web_search" | "x_search"): boolean {
+  if (!tool || typeof tool !== "object" || Array.isArray(tool)) return false;
+  const configuredType = (tool as Record<string, unknown>).type;
+  if (typeof configuredType !== "string") return false;
+  return configuredType === toolType
+    || (toolType === "web_search" && configuredType.startsWith("web_search_"));
+}
+
+function hasHostedSearchTool(value: unknown, toolType: "web_search" | "x_search"): boolean {
+  return Array.isArray(value) && value.some((tool) => isHostedSearchTool(tool, toolType));
+}
+
+function providerRequestOptionEnabled(
+  option: ProviderRequestOption,
+  extraBody: Record<string, unknown>,
+): boolean {
+  if (option.kind === "priority") return extraBody.service_tier === "priority";
+  if (Object.prototype.hasOwnProperty.call(extraBody, "tools")) {
+    return hasHostedSearchTool(extraBody.tools, option.toolType!);
+  }
+  return option.defaultEnabled === true;
+}
+
+function updateProviderRequestOption(
+  option: ProviderRequestOption,
+  enabled: boolean,
+  form: ProviderForm,
+): Partial<ProviderForm> {
+  const extraBody = { ...(parseProviderExtraBody(form.extraBody) ?? {}) };
+  if (option.kind === "priority") {
+    if (enabled) extraBody.service_tier = "priority";
+    else if (extraBody.service_tier === "priority") delete extraBody.service_tier;
+  } else {
+    const toolType = option.toolType!;
+    const tools = Array.isArray(extraBody.tools)
+      ? extraBody.tools.filter((tool) => !isHostedSearchTool(tool, toolType))
+      : [];
+    if (enabled) tools.push({ type: toolType });
+    if (tools.length || option.defaultEnabled) {
+      extraBody.tools = tools;
+    } else {
+      delete extraBody.tools;
+    }
+  }
+  return {
+    extraBody: providerJsonValue(extraBody),
+    ...(option.forceResponses && enabled
+      ? { apiType: "responses" as const }
+      : {}),
+  };
+}
+
 function providerFormFromRow(
   provider: SettingsPayload["providers"][number],
 ): ProviderForm {
@@ -3852,6 +3962,62 @@ function ModelAdvancedFields({
   );
 }
 
+function ProviderRequestOptions({
+  providerName,
+  form,
+  onChange,
+}: {
+  providerName: string;
+  form: ProviderForm;
+  onChange: (value: Partial<ProviderForm>) => void;
+}) {
+  const { t } = useTranslation();
+  const options = PROVIDER_REQUEST_OPTIONS[providerName] ?? [];
+  if (options.length === 0) return null;
+  const extraBody = parseProviderExtraBody(form.extraBody) ?? {};
+
+  const tx = (key: string, fallback: string) => t(key, { defaultValue: fallback });
+
+  return (
+    <div className="overflow-hidden rounded-[18px] border border-border/45 bg-background/75">
+      {options.map((option, index) => {
+        const title = tx(option.titleKey, option.title);
+        const Icon = option.kind === "priority" ? Zap : Globe2;
+        const checked = providerRequestOptionEnabled(option, extraBody);
+        return (
+          <div
+            key={option.titleKey}
+            className={cn(
+              "flex items-center justify-between gap-4 px-4 py-3",
+              index > 0 && "border-t border-border/45",
+            )}
+          >
+            <div className="flex min-w-0 items-start gap-3">
+              <span className="mt-0.5 flex h-8 w-8 shrink-0 items-center justify-center rounded-full bg-muted/70 text-muted-foreground">
+                <Icon className="h-4 w-4" aria-hidden />
+              </span>
+              <div className="min-w-0">
+                <p className="text-[13px] font-semibold text-foreground">{title}</p>
+                <p className="mt-0.5 text-[12px] leading-5 text-muted-foreground">
+                  {tx(option.helpKey, option.help)}
+                </p>
+              </div>
+            </div>
+            <ToggleButton
+              checked={checked}
+              onChange={(enabled) => onChange(
+                updateProviderRequestOption(option, enabled, form),
+              )}
+              ariaLabel={title}
+              label={checked ? tx("settings.values.on", "On") : tx("settings.values.off", "Off")}
+            />
+          </div>
+        );
+      })}
+    </div>
+  );
+}
+
 function ProviderAdvancedOptions({
   fields,
   form,
@@ -4065,11 +4231,11 @@ function ProviderAdvancedOptions({
               </label>
             ) : null}
           </div>
-          {footer ? (
+        </div>
-            <div className="mt-3 flex items-center justify-end gap-2 border-t border-border/45 pt-3">
+      ) : null}
-              {footer}
+      {footer ? (
-            </div>
+        <div className="flex items-center justify-end gap-2 border-t border-border/45 py-3">
-          ) : null}
+          {footer}
         </div>
       ) : null}
     </div>
@@ -4320,6 +4486,11 @@ function ProvidersSettings({
                     </Button>
                   </div>
                 </div>
+                <ProviderRequestOptions
+                  providerName={provider.name}
+                  form={form}
+                  onChange={(value) => onChangeProviderForm(provider.name, value)}
+                />
                 {supportsOauthAdvancedSettings ? (
                   <ProviderAdvancedOptions
                     fields={advancedFields}
@@ -4445,6 +4616,11 @@ function ProvidersSettings({
                     className="h-9 rounded-full text-[13px]"
                   />
                 </label>
+                <ProviderRequestOptions
+                  providerName={provider.name}
+                  form={form}
+                  onChange={(value) => onChangeProviderForm(provider.name, value)}
+                />
                 <ProviderAdvancedOptions
                   fields={advancedFields}
                   form={form}

webui/src/components/thread/activity/activity-message-model.ts

@@ -1,6 +1,7 @@
 import {
   canonicalToolTrace,
   mergeToolProgressEvents,
+  mergeToolProgressTraceLines,
   mergeUniqueToolTraceLines,
 } from "@/lib/tool-traces";
 import type { UIMediaAttachment, UIMessage } from "@/lib/types";
@@ -56,8 +57,15 @@ function canMergeAdjacentProgress(
 }
 
 function mergeTraceMessages(previous: UIMessage, incoming: UIMessage): UIMessage {
-  const traces = mergeUniqueToolTraceLines(messageTraces(previous), messageTraces(incoming)).traces;
   const toolEvents = mergeToolProgressEvents(previous.toolEvents, incoming.toolEvents ?? []);
+  const traces = incoming.toolEvents?.length
+    ? mergeToolProgressTraceLines(
+        messageTraces(previous),
+        previous.toolEvents,
+        messageTraces(incoming),
+        incoming.toolEvents ?? [],
+      )
+    : mergeUniqueToolTraceLines(messageTraces(previous), messageTraces(incoming)).traces;
   const fileEdits = [...(previous.fileEdits ?? []), ...(incoming.fileEdits ?? [])];
   const media = uniqueMedia([...(previous.media ?? []), ...(incoming.media ?? [])]);
 

webui/src/components/thread/activity/trace-activity-model.ts

@@ -27,7 +27,7 @@ export function describeTraceLine(
     !!parsedUrl && /\b(fetch(?:ing|ed)?|read(?:ing)?|opened?|opening)\b/i.test(trimmed);
 
   if (/search/i.test(name)) {
-    const query = traceFieldFromArgs(args, ["query", "q", "text"]) || args || trimmed;
+    const query = traceFieldFromArgs(args, ["query", "q", "text"]) || args;
     return {
       kind: "search",
       label: presentWebSearchAction(query, status, name === "x_search" ? "x" : "web"),

webui/src/components/thread/activity/web-search-model.ts

@@ -105,7 +105,7 @@ export function presentWebSearchAction(
   if (target === "x") {
     return queryTarget ? `${verb} X · ${queryTarget}` : `${verb} X`;
   }
-  return queryTarget ? `${verb} ${queryTarget}` : verb;
+  return queryTarget ? `${verb} ${queryTarget}` : `${verb} the web`;
 }
 
 function mergeWebSearchRun(

webui/src/hooks/useNanobotStream.ts

@@ -4,7 +4,7 @@ import { useClient } from "@/providers/ClientProvider";
 import { toMediaAttachment } from "@/lib/media";
 import {
   mergeToolProgressEvents,
-  mergeUniqueToolTraceLines,
+  mergeToolProgressTraceLines,
   normalizeToolProgressEvents,
   toolTraceLinesFromEvents,
 } from "@/lib/tool-traces";
@@ -1213,18 +1213,24 @@ export function useNanobotStream(
                 : last.content
                   ? [last.content]
                   : [];
+              const mergedEvents = visibleStructuredEvents.length > 0
+                ? mergeToolProgressEvents(last.toolEvents, visibleStructuredEvents)
+                : last.toolEvents;
               const mergedLines = visibleStructuredEvents.length > 0
-                ? mergeUniqueToolTraceLines(previousTraces, structuredLines)
+                ? mergeToolProgressTraceLines(
+                    previousTraces,
+                    last.toolEvents,
+                    structuredLines,
+                    visibleStructuredEvents,
+                  )
                 : null;
               const merged: UIMessage = {
                 ...last,
-                traces: mergedLines ? mergedLines.traces : [...previousTraces, ...lines],
+                traces: mergedLines ?? [...previousTraces, ...lines],
                 content: mergedLines
-                  ? mergedLines.traces[mergedLines.traces.length - 1]
+                  ? mergedLines[mergedLines.length - 1]
                   : lines[lines.length - 1],
-                toolEvents: visibleStructuredEvents.length
+                toolEvents: mergedEvents,
-                  ? mergeToolProgressEvents(last.toolEvents, visibleStructuredEvents)
-                  : last.toolEvents,
                 activitySegmentId: last.activitySegmentId ?? segmentId,
                 ...turn,
               };

webui/src/i18n/locales/en/common.json

@@ -514,7 +514,15 @@
       "profile": "Profile",
       "extraHeaders": "Extra headers",
       "extraBody": "Extra body",
-      "extraQuery": "Extra query"
+      "extraQuery": "Extra query",
+      "capabilityFastMode": "Fast mode",
+      "capabilityFastModeHelp": "Use OpenAI's priority service tier for faster responses. This consumes credits faster.",
+      "capabilityDeepSeekSearch": "DeepSeek web search",
+      "capabilityDeepSeekSearchHelp": "Let DeepSeek V4 Flash search the web through its Responses API. Search activity appears in chat.",
+      "capabilityXSearch": "X Search",
+      "capabilityXSearchHelp": "Allow supported Grok models to use xAI-hosted X Search. Search activity appears in chat.",
+      "capabilityOpenAISearch": "OpenAI web search",
+      "capabilityOpenAISearchHelp": "Allow compatible Responses API models to search the web. Search activity appears in chat."
     },
     "legal": {
       "thirdPartyBrands": "Product names, logos, and brands are property of their respective owners. Use is for identification only and does not imply endorsement."

webui/src/i18n/locales/es/common.json

@@ -365,7 +365,15 @@
       "profile": "Perfil",
       "extraHeaders": "Cabeceras adicionales",
       "extraBody": "Cuerpo adicional",
-      "extraQuery": "Parámetros de consulta adicionales"
+      "extraQuery": "Parámetros de consulta adicionales",
+      "capabilityFastMode": "Modo rápido",
+      "capabilityFastModeHelp": "Usa el nivel de servicio prioritario de OpenAI para responder más rápido. Esto consume créditos más deprisa.",
+      "capabilityDeepSeekSearch": "Búsqueda web de DeepSeek",
+      "capabilityDeepSeekSearchHelp": "Permite que DeepSeek V4 Flash busque en la web mediante Responses API. La actividad de búsqueda aparece en el chat.",
+      "capabilityXSearch": "Búsqueda en X",
+      "capabilityXSearchHelp": "Permite que los modelos Grok compatibles usen X Search alojado por xAI. La actividad de búsqueda aparece en el chat.",
+      "capabilityOpenAISearch": "Búsqueda web de OpenAI",
+      "capabilityOpenAISearchHelp": "Permite que los modelos compatibles con Responses API busquen en la web. La actividad de búsqueda aparece en el chat."
     },
     "image": {
       "selectProvider": "Seleccionar proveedor",

webui/src/i18n/locales/fr/common.json

@@ -365,7 +365,15 @@
       "profile": "Profil",
       "extraHeaders": "En-têtes supplémentaires",
       "extraBody": "Corps supplémentaire",
-      "extraQuery": "Paramètres de requête supplémentaires"
+      "extraQuery": "Paramètres de requête supplémentaires",
+      "capabilityFastMode": "Mode rapide",
+      "capabilityFastModeHelp": "Utilise le niveau de service prioritaire d’OpenAI pour accélérer les réponses. Les crédits sont consommés plus rapidement.",
+      "capabilityDeepSeekSearch": "Recherche web DeepSeek",
+      "capabilityDeepSeekSearchHelp": "Permet à DeepSeek V4 Flash de rechercher sur le web via l’API Responses. L’activité de recherche apparaît dans le chat.",
+      "capabilityXSearch": "Recherche sur X",
+      "capabilityXSearchHelp": "Permet aux modèles Grok compatibles d’utiliser X Search hébergé par xAI. L’activité de recherche apparaît dans le chat.",
+      "capabilityOpenAISearch": "Recherche web OpenAI",
+      "capabilityOpenAISearchHelp": "Permet aux modèles compatibles avec l’API Responses de rechercher sur le web. L’activité de recherche apparaît dans le chat."
     },
     "image": {
       "selectProvider": "Choisir un fournisseur",

webui/src/i18n/locales/id/common.json

@@ -365,7 +365,15 @@
       "profile": "Profil",
       "extraHeaders": "Header tambahan",
       "extraBody": "Body tambahan",
-      "extraQuery": "Parameter kueri tambahan"
+      "extraQuery": "Parameter kueri tambahan",
+      "capabilityFastMode": "Mode cepat",
+      "capabilityFastModeHelp": "Gunakan tingkat layanan prioritas OpenAI agar respons lebih cepat. Kredit akan terpakai lebih cepat.",
+      "capabilityDeepSeekSearch": "Pencarian web DeepSeek",
+      "capabilityDeepSeekSearchHelp": "Izinkan DeepSeek V4 Flash mencari di web melalui Responses API. Aktivitas pencarian ditampilkan di chat.",
+      "capabilityXSearch": "Pencarian X",
+      "capabilityXSearchHelp": "Izinkan model Grok yang didukung menggunakan X Search yang dihosting xAI. Aktivitas pencarian ditampilkan di chat.",
+      "capabilityOpenAISearch": "Pencarian web OpenAI",
+      "capabilityOpenAISearchHelp": "Izinkan model yang kompatibel dengan Responses API mencari di web. Aktivitas pencarian ditampilkan di chat."
     },
     "image": {
       "selectProvider": "Pilih penyedia",

webui/src/i18n/locales/ja/common.json

@@ -365,7 +365,15 @@
       "profile": "プロファイル",
       "extraHeaders": "追加ヘッダー",
       "extraBody": "追加ボディ",
-      "extraQuery": "追加クエリ"
+      "extraQuery": "追加クエリ",
+      "capabilityFastMode": "高速モード",
+      "capabilityFastModeHelp": "OpenAI の優先サービス階層を使用して応答を高速化します。クレジットの消費も速くなります。",
+      "capabilityDeepSeekSearch": "DeepSeek ウェブ検索",
+      "capabilityDeepSeekSearchHelp": "DeepSeek V4 Flash が Responses API 経由でウェブを検索できるようにします。検索状況はチャットに表示されます。",
+      "capabilityXSearch": "X 検索",
+      "capabilityXSearchHelp": "対応する Grok モデルが xAI ホストの X Search を使用できるようにします。検索状況はチャットに表示されます。",
+      "capabilityOpenAISearch": "OpenAI ウェブ検索",
+      "capabilityOpenAISearchHelp": "Responses API 対応モデルがウェブを検索できるようにします。検索状況はチャットに表示されます。"
     },
     "image": {
       "selectProvider": "プロバイダーを選択",

webui/src/i18n/locales/ko/common.json

@@ -365,7 +365,15 @@
       "profile": "프로필",
       "extraHeaders": "추가 헤더",
       "extraBody": "추가 본문",
-      "extraQuery": "추가 쿼리"
+      "extraQuery": "추가 쿼리",
+      "capabilityFastMode": "고속 모드",
+      "capabilityFastModeHelp": "OpenAI 우선 서비스 등급을 사용해 더 빠르게 응답합니다. 크레딧도 더 빠르게 소모됩니다.",
+      "capabilityDeepSeekSearch": "DeepSeek 웹 검색",
+      "capabilityDeepSeekSearchHelp": "DeepSeek V4 Flash가 Responses API를 통해 웹을 검색하도록 허용합니다. 검색 활동은 채팅에 표시됩니다.",
+      "capabilityXSearch": "X 검색",
+      "capabilityXSearchHelp": "지원되는 Grok 모델이 xAI 호스팅 X Search를 사용하도록 허용합니다. 검색 활동은 채팅에 표시됩니다.",
+      "capabilityOpenAISearch": "OpenAI 웹 검색",
+      "capabilityOpenAISearchHelp": "Responses API 호환 모델이 웹을 검색하도록 허용합니다. 검색 활동은 채팅에 표시됩니다."
     },
     "image": {
       "selectProvider": "제공자 선택",

webui/src/i18n/locales/pt-BR/common.json

@@ -514,7 +514,15 @@
       "profile": "Perfil",
       "extraHeaders": "Cabeçalhos adicionais",
       "extraBody": "Corpo adicional",
-      "extraQuery": "Parâmetros de consulta adicionais"
+      "extraQuery": "Parâmetros de consulta adicionais",
+      "capabilityFastMode": "Modo rápido",
+      "capabilityFastModeHelp": "Usa o nível de serviço prioritário da OpenAI para respostas mais rápidas. Isso consome créditos mais rapidamente.",
+      "capabilityDeepSeekSearch": "Pesquisa web do DeepSeek",
+      "capabilityDeepSeekSearchHelp": "Permite que o DeepSeek V4 Flash pesquise na web pela Responses API. A atividade de pesquisa aparece no chat.",
+      "capabilityXSearch": "Pesquisa no X",
+      "capabilityXSearchHelp": "Permite que modelos Grok compatíveis usem o X Search hospedado pela xAI. A atividade de pesquisa aparece no chat.",
+      "capabilityOpenAISearch": "Pesquisa web da OpenAI",
+      "capabilityOpenAISearchHelp": "Permite que modelos compatíveis com a Responses API pesquisem na web. A atividade de pesquisa aparece no chat."
     },
     "legal": {
       "thirdPartyBrands": "Nomes de produtos, logotipos e marcas são propriedades de seus respectivos donos. O uso é apenas para identificação e não implica endosso."

webui/src/i18n/locales/vi/common.json

@@ -365,7 +365,15 @@
       "profile": "Hồ sơ",
       "extraHeaders": "Header bổ sung",
       "extraBody": "Body bổ sung",
-      "extraQuery": "Tham số truy vấn bổ sung"
+      "extraQuery": "Tham số truy vấn bổ sung",
+      "capabilityFastMode": "Chế độ nhanh",
+      "capabilityFastModeHelp": "Dùng tầng dịch vụ ưu tiên của OpenAI để phản hồi nhanh hơn. Tín dụng cũng sẽ được tiêu thụ nhanh hơn.",
+      "capabilityDeepSeekSearch": "Tìm kiếm web DeepSeek",
+      "capabilityDeepSeekSearchHelp": "Cho phép DeepSeek V4 Flash tìm kiếm trên web qua Responses API. Hoạt động tìm kiếm được hiển thị trong cuộc trò chuyện.",
+      "capabilityXSearch": "Tìm kiếm X",
+      "capabilityXSearchHelp": "Cho phép các mô hình Grok được hỗ trợ dùng X Search do xAI lưu trữ. Hoạt động tìm kiếm được hiển thị trong cuộc trò chuyện.",
+      "capabilityOpenAISearch": "Tìm kiếm web OpenAI",
+      "capabilityOpenAISearchHelp": "Cho phép các mô hình tương thích với Responses API tìm kiếm trên web. Hoạt động tìm kiếm được hiển thị trong cuộc trò chuyện."
     },
     "image": {
       "selectProvider": "Chọn nhà cung cấp",

webui/src/i18n/locales/zh-CN/common.json

@@ -514,7 +514,15 @@
       "profile": "配置档案",
       "extraHeaders": "额外请求头",
       "extraBody": "额外请求体",
-      "extraQuery": "额外查询参数"
+      "extraQuery": "额外查询参数",
+      "capabilityFastMode": "Fast 模式",
+      "capabilityFastModeHelp": "通过 OpenAI 优先服务层获取更快响应，会更快消耗额度。",
+      "capabilityDeepSeekSearch": "DeepSeek 联网搜索",
+      "capabilityDeepSeekSearchHelp": "允许 DeepSeek V4 Flash 通过 Responses API 搜索网络，并在对话中展示搜索过程。",
+      "capabilityXSearch": "X 搜索",
+      "capabilityXSearchHelp": "允许受支持的 Grok 模型使用 xAI 托管的 X Search，并在对话中展示搜索过程。",
+      "capabilityOpenAISearch": "OpenAI 联网搜索",
+      "capabilityOpenAISearchHelp": "允许兼容 Responses API 的模型搜索网络，并在对话中展示搜索过程。"
     },
     "legal": {
       "thirdPartyBrands": "产品名称、Logo 和品牌归各自所有者所有；此处仅用于识别，不代表背书或合作。"

webui/src/i18n/locales/zh-TW/common.json

@@ -365,7 +365,15 @@
       "profile": "設定檔",
       "extraHeaders": "額外請求標頭",
       "extraBody": "額外請求本文",
-      "extraQuery": "額外查詢參數"
+      "extraQuery": "額外查詢參數",
+      "capabilityFastMode": "Fast 模式",
+      "capabilityFastModeHelp": "透過 OpenAI 優先服務層取得更快回應，也會更快消耗額度。",
+      "capabilityDeepSeekSearch": "DeepSeek 網路搜尋",
+      "capabilityDeepSeekSearchHelp": "允許 DeepSeek V4 Flash 透過 Responses API 搜尋網路，並在對話中顯示搜尋過程。",
+      "capabilityXSearch": "X 搜尋",
+      "capabilityXSearchHelp": "允許支援的 Grok 模型使用 xAI 託管的 X Search，並在對話中顯示搜尋過程。",
+      "capabilityOpenAISearch": "OpenAI 網路搜尋",
+      "capabilityOpenAISearchHelp": "允許相容 Responses API 的模型搜尋網路，並在對話中顯示搜尋過程。"
     },
     "image": {
       "selectProvider": "選擇供應商",

webui/src/lib/tool-traces.ts

@@ -16,7 +16,10 @@ export function formatToolCallTrace(call: unknown): string | null {
   if (!name) return null;
   const args = item.function?.arguments ?? item.arguments;
   if (typeof args === "string" && args.trim()) return `${name}(${args})`;
-  if (args && typeof args === "object") return `${name}(${JSON.stringify(args)})`;
+  if (args && typeof args === "object") {
+    const serialized = JSON.stringify(args);
+    return serialized === "{}" || serialized === "[]" ? `${name}()` : `${name}(${serialized})`;
+  }
   return `${name}()`;
 }
 
@@ -116,3 +119,24 @@ export function mergeUniqueToolTraceLines(
   }
   return { traces, added };
 }
+
+export function mergeToolProgressTraceLines(
+  previousTraces: string[],
+  previousEvents: ToolProgressEvent[] | undefined,
+  incomingTraces: string[],
+  incomingEvents: ToolProgressEvent[],
+): string[] {
+  const mergedEvents = mergeToolProgressEvents(previousEvents, incomingEvents);
+  const candidates = mergeUniqueToolTraceLines(previousTraces, incomingTraces).traces;
+  const eventTraceKeys = new Set([
+    ...toolTraceLinesFromEvents(previousEvents),
+    ...toolTraceLinesFromEvents(incomingEvents),
+  ].map(canonicalToolTrace));
+  const nonEventTraces = candidates.filter(
+    (line) => !eventTraceKeys.has(canonicalToolTrace(line)),
+  );
+  return mergeUniqueToolTraceLines(
+    nonEventTraces,
+    toolTraceLinesFromEvents(mergedEvents),
+  ).traces;
+}

webui/src/tests/agent-activity-cluster.test.tsx

@@ -1165,12 +1165,13 @@ describe("AgentActivityCluster", () => {
             id: "search-start",
             role: "tool",
             kind: "trace",
-            content: line,
+            content: "web_search()",
-            traces: [line],
+            traces: ["web_search()"],
             toolEvents: [{
               phase: "start",
+              call_id: "hosted-search-1",
               name: "web_search",
-              arguments: { query: "site:linkedin.com/company Evomap startup" },
+              arguments: {},
             }],
             createdAt: 1,
           },
@@ -1182,6 +1183,7 @@ describe("AgentActivityCluster", () => {
             traces: [line],
             toolEvents: [{
               phase: "error",
+              call_id: "hosted-search-1",
               name: "web_search",
               arguments: { query: "site:linkedin.com/company Evomap startup" },
               error: "Search provider rate limited the request",

webui/src/tests/api.test.ts

@@ -580,7 +580,7 @@ describe("webui API helpers", () => {
     await updateProviderSettings("tok", {
       provider: "xai_grok",
       proxy: "http://127.0.0.1:7890",
-      extraBody: '{"service_tier":"priority"}',
+      extraBody: '{"tools":[]}',
     });
 
     expect(fetch).toHaveBeenCalledWith(
@@ -590,7 +590,7 @@ describe("webui API helpers", () => {
           Authorization: "Bearer tok",
           "X-Nanobot-Provider-Values": encodeURIComponent(JSON.stringify({
             proxy: "http://127.0.0.1:7890",
-            extraBody: '{"service_tier":"priority"}',
+            extraBody: '{"tools":[]}',
           })),
         },
       }),

webui/src/tests/settings-view.test.tsx

@@ -3101,6 +3101,200 @@ describe("SettingsView Apps catalog", () => {
     );
   });
 
+  it("maps provider request switches to raw extraBody fields", async () => {
+    const base = settingsPayload();
+    const providers: SettingsPayload["providers"] = [
+      {
+        name: "xai_grok",
+        label: "xAI Grok",
+        configured: true,
+        auth_type: "oauth",
+        api_key_required: false,
+        oauth_account: "grok@example.com",
+        oauth_login_supported: true,
+        advanced_fields: ["extra_body", "proxy"],
+        extra_body: null,
+      },
+      {
+        name: "openai_codex",
+        label: "OpenAI Codex",
+        configured: true,
+        auth_type: "oauth",
+        api_key_required: false,
+        oauth_account: "codex@example.com",
+        oauth_login_supported: true,
+        advanced_fields: ["extra_body", "proxy"],
+        extra_body: null,
+      },
+      {
+        name: "deepseek",
+        label: "DeepSeek",
+        configured: true,
+        api_key_required: true,
+        api_key_hint: "deep••••test",
+        api_base: "https://api.deepseek.com",
+        advanced_fields: ["extra_body"],
+        extra_body: null,
+      },
+      {
+        name: "openai",
+        label: "OpenAI",
+        configured: true,
+        api_key_required: true,
+        api_key_hint: "sk-••••test",
+        api_base: "https://api.openai.com/v1",
+        api_type: "auto",
+        advanced_fields: ["api_type", "extra_body"],
+        extra_body: null,
+      },
+    ];
+    const payload: SettingsPayload = { ...base, providers };
+    const fetchMock = vi.fn(async (...args: [RequestInfo | URL, RequestInit?]) => {
+      const [input] = args;
+      const url = String(input);
+      if (url === "/api/settings") return jsonResponse(payload);
+      if (url.startsWith("/api/settings/provider/update?")) return jsonResponse(payload);
+      if (url === "/api/settings/cli-apps") {
+        return jsonResponse({ apps: [], installed_count: 0 });
+      }
+      if (url === "/api/settings/mcp-presets") {
+        return jsonResponse({ presets: [], installed_count: 0 });
+      }
+      return jsonResponse({});
+    });
+    vi.stubGlobal("fetch", fetchMock);
+
+    renderSettingsView({ initialSection: "models", initialSettings: payload });
+
+    fireEvent.click(screen.getByRole("button", { name: "xAI Grok" }));
+    const xSearch = screen.getByRole("switch", { name: "X Search" });
+    expect(xSearch).toHaveAttribute("aria-checked", "true");
+    fireEvent.click(xSearch);
+    fireEvent.click(screen.getByRole("button", { name: "Save provider" }));
+    await waitFor(() => expect(fetchMock).toHaveBeenCalledWith(
+      "/api/settings/provider/update?provider=xai_grok",
+      expect.anything(),
+    ));
+    await waitFor(() => expect(
+      screen.getByRole("button", { name: "Save provider" }),
+    ).toBeEnabled());
+
+    fireEvent.click(screen.getByRole("button", { name: "xAI Grok" }));
+    fireEvent.click(screen.getByRole("button", { name: "OpenAI Codex" }));
+    fireEvent.click(screen.getByRole("switch", { name: "Fast mode" }));
+    fireEvent.click(screen.getByRole("button", { name: "Save provider" }));
+    await waitFor(() => expect(fetchMock).toHaveBeenCalledWith(
+      "/api/settings/provider/update?provider=openai_codex",
+      expect.anything(),
+    ));
+    await waitFor(() => expect(
+      screen.getByRole("button", { name: "Save provider" }),
+    ).toBeEnabled());
+
+    fireEvent.click(screen.getByRole("button", { name: "OpenAI Codex" }));
+    fireEvent.click(screen.getByRole("button", { name: /^DeepSeek/ }));
+    expect(screen.getByText(/DeepSeek V4 Flash/)).toBeInTheDocument();
+    const deepSeekSearch = screen.getByRole("switch", { name: "DeepSeek web search" });
+    expect(deepSeekSearch).toHaveAttribute("aria-checked", "true");
+    fireEvent.click(deepSeekSearch);
+    fireEvent.click(screen.getByRole("button", { name: "Save provider" }));
+    await waitFor(() => expect(
+      screen.queryByRole("switch", { name: "DeepSeek web search" }),
+    ).not.toBeInTheDocument());
+
+    fireEvent.click(screen.getByRole("button", { name: /^OpenAI https:/ }));
+    fireEvent.click(screen.getByRole("switch", { name: "OpenAI web search" }));
+    fireEvent.click(screen.getByRole("button", { name: "Save provider" }));
+    await waitFor(() => expect(
+      screen.queryByRole("switch", { name: "OpenAI web search" }),
+    ).not.toBeInTheDocument());
+
+    await waitFor(() => {
+      const requestUpdates = fetchMock.mock.calls
+        .filter(([input]) => String(input).startsWith("/api/settings/provider/update?"))
+        .map(([input, init]) => {
+          const provider = new URLSearchParams(String(input).split("?")[1]).get("provider");
+          const headers = init?.headers as Record<string, string>;
+          const values = JSON.parse(decodeURIComponent(
+            headers["X-Nanobot-Provider-Values"],
+          )) as { apiType?: string; extraBody?: string };
+          return [provider, {
+            ...(values.apiType ? { apiType: values.apiType } : {}),
+            extraBody: JSON.parse(values.extraBody ?? "{}"),
+          }] as const;
+        });
+      expect(requestUpdates).toEqual([
+        ["xai_grok", { extraBody: { tools: [] } }],
+        ["openai_codex", { extraBody: { service_tier: "priority" } }],
+        ["deepseek", { extraBody: { tools: [] } }],
+        ["openai", {
+          apiType: "responses",
+          extraBody: { tools: [{ type: "web_search" }] },
+        }],
+      ]);
+    });
+  });
+
+  it("recognizes and removes versioned web search tools without losing raw settings", async () => {
+    const base = settingsPayload();
+    const payload: SettingsPayload = {
+      ...base,
+      providers: [{
+        name: "openai",
+        label: "OpenAI",
+        configured: true,
+        api_key_required: true,
+        api_key_hint: "sk-••••test",
+        api_base: "https://api.openai.com/v1",
+        api_type: "auto",
+        advanced_fields: ["api_type", "extra_body"],
+        extra_body: {
+          metadata: { owner: "legacy-config" },
+          tools: [
+            { type: "web_search_preview", search_context_size: "medium" },
+            { type: "file_search", vector_store_ids: ["vs_legacy"] },
+          ],
+        },
+      }],
+    };
+    const fetchMock = vi.fn(async (input: RequestInfo | URL) => {
+      const url = String(input);
+      if (url === "/api/settings") return jsonResponse(payload);
+      if (url.startsWith("/api/settings/provider/update?")) return jsonResponse(payload);
+      if (url === "/api/settings/cli-apps") {
+        return jsonResponse({ apps: [], installed_count: 0 });
+      }
+      if (url === "/api/settings/mcp-presets") {
+        return jsonResponse({ presets: [], installed_count: 0 });
+      }
+      return jsonResponse({});
+    });
+    vi.stubGlobal("fetch", fetchMock);
+
+    renderSettingsView({ initialSection: "models", initialSettings: payload });
+
+    fireEvent.click(await screen.findByRole("button", { name: /^OpenAI https:/ }));
+    const searchSwitch = screen.getByRole("switch", { name: "OpenAI web search" });
+    expect(searchSwitch).toHaveAttribute("aria-checked", "true");
+    fireEvent.click(searchSwitch);
+    fireEvent.click(screen.getByRole("button", { name: "Save provider" }));
+
+    await waitFor(() => {
+      const updateCall = fetchMock.mock.calls.find(
+        ([input]) => String(input).startsWith("/api/settings/provider/update?"),
+      );
+      expect(updateCall).toBeTruthy();
+      const headers = updateCall?.[1]?.headers as Record<string, string>;
+      const values = JSON.parse(decodeURIComponent(
+        headers["X-Nanobot-Provider-Values"],
+      )) as { extraBody: string };
+      expect(JSON.parse(values.extraBody)).toEqual({
+        metadata: { owner: "legacy-config" },
+        tools: [{ type: "file_search", vector_store_ids: ["vs_legacy"] }],
+      });
+    });
+  });
+
   it("creates a custom provider with folded advanced request settings", async () => {
     const base = settingsPayload();
     let payload: SettingsPayload = {

webui/src/tests/tool-traces.test.ts

@@ -2,6 +2,7 @@ import { describe, expect, it } from "vitest";
 
 import {
   canonicalToolTrace,
+  mergeToolProgressTraceLines,
   mergeUniqueToolTraceLines,
 } from "@/lib/tool-traces";
 
@@ -26,4 +27,18 @@ describe("tool trace identity", () => {
       added: true,
     });
   });
+
+  it("replaces an empty streaming placeholder when hosted search arguments arrive", () => {
+    expect(mergeToolProgressTraceLines(
+      ["web_search()"],
+      [{ phase: "start", call_id: "ws-1", name: "web_search", arguments: {} }],
+      ['web_search({"query":"nanobot news"})'],
+      [{
+        phase: "end",
+        call_id: "ws-1",
+        name: "web_search",
+        arguments: { query: "nanobot news" },
+      }],
+    )).toEqual(['web_search({"query":"nanobot news"})']);
+  });
 });

webui/src/tests/trace-activity-model.test.ts

@@ -8,6 +8,10 @@ function describeTrace(line: string, status: GenericToolStatus = "done") {
 }
 
 describe("trace activity semantics", () => {
+  it("uses readable copy when a hosted search query is unavailable", () => {
+    expect(describeTrace("web_search()").label).toBe("Searched the web");
+  });
+
   it.each([
     ['web_search({"query":"nanobot latest release"})', "done", "Searched nanobot latest release", ""],
     ['web_fetch({"url":"https://example.com/docs?token=private"})', "done", "Read", "example.com/docs"],

webui/src/tests/useNanobotStream.test.tsx

@@ -610,6 +610,51 @@ describe("useNanobotStream", () => {
     ]);
   });
 
+  it("replaces a hosted search placeholder when its query arrives", () => {
+    const fake = fakeClient();
+    const { result } = renderHook(() => useNanobotStream("chat-hosted-search", EMPTY_MESSAGES), {
+      wrapper: wrap(fake.client),
+    });
+
+    act(() => {
+      fake.emit("chat-hosted-search", {
+        event: "message",
+        chat_id: "chat-hosted-search",
+        text: "web_search()",
+        kind: "tool_hint",
+        tool_events: [{
+          phase: "start",
+          call_id: "ws-1",
+          name: "web_search",
+          arguments: {},
+        }],
+      });
+      fake.emit("chat-hosted-search", {
+        event: "message",
+        chat_id: "chat-hosted-search",
+        text: "",
+        kind: "progress",
+        tool_events: [{
+          phase: "end",
+          call_id: "ws-1",
+          name: "web_search",
+          arguments: { query: "nanobot news" },
+          result: { status: "completed" },
+        }],
+      });
+    });
+
+    expect(result.current.messages).toHaveLength(1);
+    expect(result.current.messages[0].traces).toEqual([
+      'web_search({"query":"nanobot news"})',
+    ]);
+    expect(result.current.messages[0].toolEvents).toMatchObject([{
+      phase: "end",
+      call_id: "ws-1",
+      arguments: { query: "nanobot news" },
+    }]);
+  });
+
   it("keeps phase updates when a tool event trace line is deduped", () => {
     const fake = fakeClient();
     const { result } = renderHook(() => useNanobotStream("chat-tool-phase", EMPTY_MESSAGES), {