[pull] master from AstrBotDevs:master

astrbot/core/astr_main_agent.py

@@ -77,6 +77,8 @@
     BaiduWebSearchTool,
     BochaWebSearchTool,
     BraveWebSearchTool,
+    FirecrawlExtractWebPageTool,
+    FirecrawlWebSearchTool,
     TavilyExtractWebPageTool,
     TavilyWebSearchTool,
     normalize_legacy_web_search_config,
@@ -1047,6 +1049,9 @@ async def _apply_web_search_tools(
         req.func_tool.add_tool(tool_mgr.get_builtin_tool(BochaWebSearchTool))
     elif provider == "brave":
         req.func_tool.add_tool(tool_mgr.get_builtin_tool(BraveWebSearchTool))
+    elif provider == "firecrawl":
+        req.func_tool.add_tool(tool_mgr.get_builtin_tool(FirecrawlWebSearchTool))
+        req.func_tool.add_tool(tool_mgr.get_builtin_tool(FirecrawlExtractWebPageTool))
     elif provider == "baidu_ai_search":
         req.func_tool.add_tool(tool_mgr.get_builtin_tool(BaiduWebSearchTool))
 

astrbot/core/config/default.py

@@ -3202,6 +3202,7 @@ class ChatProviderTemplate(TypedDict):
                             "baidu_ai_search",
                             "bocha",
                             "brave",
+                            "firecrawl",
                         ],
                         "condition": {
                             "provider_settings.web_search": True,
@@ -3237,6 +3238,16 @@ class ChatProviderTemplate(TypedDict):
                             "provider_settings.web_search": True,
                         },
                     },
+                    "provider_settings.websearch_firecrawl_key": {
+                        "description": "Firecrawl API Key",
+                        "type": "list",
+                        "items": {"type": "string"},
+                        "hint": "可添加多个 Key 进行轮询。",
+                        "condition": {
+                            "provider_settings.websearch_provider": "firecrawl",
+                            "provider_settings.web_search": True,
+                        },
+                    },
                     "provider_settings.websearch_baidu_app_builder_key": {
                         "description": "百度千帆智能云 APP Builder API Key",
                         "type": "string",

astrbot/core/provider/sources/openai_source.py

@@ -519,6 +519,42 @@ async def get_models(self):
         except NotFoundError as e:
             raise Exception(f"获取模型列表失败：{e}")
 
+    @staticmethod
+    def _sanitize_assistant_messages(payloads: dict) -> None:
+        """在请求发送前过滤/规范化空的 assistant 消息。
+
+        严格 API（Moonshot、DeepSeek Reasoner 等）会在 assistant 消息同时缺少
+        ``content`` 和 ``tool_calls`` 时返回 400。把 ``""`` / ``None`` / ``[]``
+        都视作空内容：无 tool_calls 时整条过滤掉；有 tool_calls 时将 content
+        设为 ``None`` 以符合 OpenAI 规范。就地修改 ``payloads["messages"]``。
+        """
+        messages = payloads.get("messages")
+        if not isinstance(messages, list):
+            return
+
+        def _is_empty(content: Any) -> bool:
+            return content is None or content == "" or content == []
+
+        cleaned: list[Any] = []
+        for idx, msg in enumerate(messages):
+            if not isinstance(msg, dict) or msg.get("role") != "assistant":
+                cleaned.append(msg)
+                continue
+
+            content = msg.get("content")
+            tool_calls = msg.get("tool_calls")
+
+            if _is_empty(content) and not tool_calls:
+                logger.warning(f"过滤第 {idx} 条空 assistant 消息 (无工具调用)")
+                continue
+
+            if _is_empty(content) and tool_calls:
+                msg["content"] = None
+
+            cleaned.append(msg)
+
+        payloads["messages"] = cleaned
+
     async def _query(self, payloads: dict, tools: ToolSet | None) -> LLMResponse:
         if tools:
             model = payloads.get("model", "").lower()
@@ -548,26 +584,7 @@ async def _query(self, payloads: dict, tools: ToolSet | None) -> LLMResponse:
 
         model = payloads.get("model", "").lower()
 
-        if "messages" in payloads and isinstance(payloads["messages"], list):
-            cleaned_messages = []
-            for idx, msg in enumerate(payloads["messages"]):
-                # 过滤空的 assistant 消息，防止严格 API（如 Moonshot）返回 400 错误
-                if msg.get("role") == "assistant":
-                    content = msg.get("content")
-                    tool_calls = msg.get("tool_calls")
-
-                    # 情况1: 空/null content 且无 tool_calls -> 过滤掉
-                    if not tool_calls and (content == "" or content is None):
-                        logger.warning(f"过滤第 {idx} 条空 assistant 消息 (无工具调用)")
-                        continue
-
-                    # 情况2: 空 content 但有 tool_calls -> 设为 None (符合 OpenAI 规范)
-                    if content == "" and tool_calls:
-                        msg["content"] = None
-
-                cleaned_messages.append(msg)
-
-            payloads["messages"] = cleaned_messages
+        self._sanitize_assistant_messages(payloads)
 
         completion = await self.client.chat.completions.create(
             **payloads,
@@ -619,6 +636,8 @@ async def _query_stream(
             del payloads[key]
         self._apply_provider_specific_extra_body_overrides(extra_body)
 
+        self._sanitize_assistant_messages(payloads)
+
         stream = await self.client.chat.completions.create(
             **payloads,
             stream=True,

astrbot/core/tools/web_search_tools.py

@@ -19,6 +19,8 @@
     "tavily_extract_web_page",
     "web_search_bocha",
     "web_search_brave",
+    "web_search_firecrawl",
+    "firecrawl_extract_web_page",
 ]
 _TAVILY_WEB_SEARCH_TOOL_CONFIG = {
     "provider_settings.web_search": True,
@@ -32,6 +34,10 @@
     "provider_settings.web_search": True,
     "provider_settings.websearch_provider": "brave",
 }
+_FIRECRAWL_WEB_SEARCH_TOOL_CONFIG = {
+    "provider_settings.web_search": True,
+    "provider_settings.websearch_provider": "firecrawl",
+}
 _BAIDU_WEB_SEARCH_TOOL_CONFIG = {
     "provider_settings.web_search": True,
     "provider_settings.websearch_provider": "baidu_ai_search",
@@ -69,6 +75,7 @@ async def get(self, provider_settings: dict) -> str:
 _TAVILY_KEY_ROTATOR = _KeyRotator("websearch_tavily_key", "Tavily")
 _BOCHA_KEY_ROTATOR = _KeyRotator("websearch_bocha_key", "BoCha")
 _BRAVE_KEY_ROTATOR = _KeyRotator("websearch_brave_key", "Brave")
+_FIRECRAWL_KEY_ROTATOR = _KeyRotator("websearch_firecrawl_key", "Firecrawl")
 
 
 def normalize_legacy_web_search_config(cfg) -> None:
@@ -91,6 +98,7 @@ def normalize_legacy_web_search_config(cfg) -> None:
         "websearch_tavily_key",
         "websearch_bocha_key",
         "websearch_brave_key",
+        "websearch_firecrawl_key",
     ):
         value = provider_settings.get(setting_name)
         if isinstance(value, str):
@@ -258,6 +266,72 @@ async def _brave_search(
             ]
 
 
+async def _firecrawl_search(
+    provider_settings: dict,
+    payload: dict,
+) -> list[SearchResult]:
+    firecrawl_key = await _FIRECRAWL_KEY_ROTATOR.get(provider_settings)
+    header = {
+        "Authorization": f"Bearer {firecrawl_key}",
+        "Content-Type": "application/json",
+    }
+    async with aiohttp.ClientSession(trust_env=True) as session:
+        async with session.post(
+            "https://api.firecrawl.dev/v2/search",
+            json=payload,
+            headers=header,
+        ) as response:
+            if response.status != 200:
+                reason = await response.text()
+                raise Exception(
+                    f"Firecrawl web search failed: {reason}, status: {response.status}",
+                )
+            data = await response.json()
+            rows = data.get("data", [])
+            if isinstance(rows, dict):
+                rows = rows.get("web", [])
+            return [
+                SearchResult(
+                    title=item.get("title", ""),
+                    url=item.get("url", ""),
+                    snippet=(
+                        item.get("description")
+                        or item.get("snippet")
+                        or item.get("markdown")
+                        or ""
+                    ),
+                )
+                for item in rows
+                if item.get("url")
+            ]
+
+
+async def _firecrawl_scrape(provider_settings: dict, payload: dict) -> dict:
+    firecrawl_key = await _FIRECRAWL_KEY_ROTATOR.get(provider_settings)
+    header = {
+        "Authorization": f"Bearer {firecrawl_key}",
+        "Content-Type": "application/json",
+    }
+    async with aiohttp.ClientSession(trust_env=True) as session:
+        async with session.post(
+            "https://api.firecrawl.dev/v2/scrape",
+            json=payload,
+            headers=header,
+        ) as response:
+            if response.status != 200:
+                reason = await response.text()
+                raise Exception(
+                    f"Firecrawl web scraper failed: {reason}, status: {response.status}",
+                )
+            data = await response.json()
+            result = data.get("data", {})
+            if not result:
+                raise ValueError(
+                    "Error: Firecrawl web scraper does not return any results."
+                )
+            return result
+
+
 async def _baidu_search(
     provider_settings: dict,
     payload: dict,
@@ -548,6 +622,124 @@ async def call(self, context, **kwargs) -> ToolExecResult:
         return _search_result_payload(results)
 
 
+@builtin_tool(config=_FIRECRAWL_WEB_SEARCH_TOOL_CONFIG)
+@pydantic_dataclass
+class FirecrawlWebSearchTool(FunctionTool[AstrAgentContext]):
+    name: str = "web_search_firecrawl"
+    description: str = (
+        "A web search tool based on Firecrawl Search API, used to retrieve web "
+        "pages related to the user's query."
+    )
+    parameters: dict = Field(
+        default_factory=lambda: {
+            "type": "object",
+            "properties": {
+                "query": {"type": "string", "description": "Required. Search query."},
+                "limit": {
+                    "type": "integer",
+                    "description": "Optional. Number of results to return. Range: 1-100. Default is 5.",
+                },
+                "location": {
+                    "type": "string",
+                    "description": "Optional. Geographic location for search results.",
+                },
+                "country": {
+                    "type": "string",
+                    "description": 'Optional. Country code for search results, for example "US" or "CN".',
+                },
+                "timeout": {
+                    "type": "integer",
+                    "description": "Optional. Request timeout in milliseconds.",
+                },
+            },
+            "required": ["query"],
+        }
+    )
+
+    async def call(self, context, **kwargs) -> ToolExecResult:
+        _, provider_settings, _ = _get_runtime(context)
+        if not provider_settings.get("websearch_firecrawl_key", []):
+            return "Error: Firecrawl API key is not configured in AstrBot."
+
+        payload = {
+            "query": kwargs["query"],
+            "limit": kwargs.get("limit", 5),
+            "sources": ["web"],
+        }
+        for key in ("location", "country", "timeout"):
+            if kwargs.get(key):
+                payload[key] = kwargs[key]
+
+        results = await _firecrawl_search(provider_settings, payload)
+        if not results:
+            return "Error: Firecrawl web searcher does not return any results."
+        return _search_result_payload(results)
+
+
+@builtin_tool(config=_FIRECRAWL_WEB_SEARCH_TOOL_CONFIG)
+@pydantic_dataclass
+class FirecrawlExtractWebPageTool(FunctionTool[AstrAgentContext]):
+    name: str = "firecrawl_extract_web_page"
+    description: str = "Extract the content of a web page using Firecrawl."
+    parameters: dict = Field(
+        default_factory=lambda: {
+            "type": "object",
+            "properties": {
+                "url": {
+                    "type": "string",
+                    "description": "Required. A URL to extract content from.",
+                },
+                "format": {
+                    "type": "string",
+                    "description": 'Optional. Output format, one of "markdown", "html", "rawHtml", "summary". Default is "markdown".',
+                },
+                "only_main_content": {
+                    "type": "boolean",
+                    "description": "Optional. Whether to extract only the main page content. Default is true.",
+                },
+                "timeout": {
+                    "type": "integer",
+                    "description": "Optional. Request timeout in milliseconds.",
+                },
+                "max_age": {
+                    "type": "integer",
+                    "description": "Optional. Maximum cache age in milliseconds.",
+                },
+            },
+            "required": ["url"],
+        }
+    )
+
+    async def call(self, context, **kwargs) -> ToolExecResult:
+        _, provider_settings, _ = _get_runtime(context)
+        if not provider_settings.get("websearch_firecrawl_key", []):
+            return "Error: Firecrawl API key is not configured in AstrBot."
+
+        url = str(kwargs.get("url", "")).strip()
+        if not url:
+            return "Error: url must be a non-empty string."
+
+        output_format = kwargs.get("format", "markdown")
+        if output_format not in ["markdown", "html", "rawHtml", "summary"]:
+            output_format = "markdown"
+
+        payload = {
+            "url": url,
+            "formats": [output_format],
+            "onlyMainContent": kwargs.get("only_main_content", True),
+        }
+        if kwargs.get("timeout"):
+            payload["timeout"] = kwargs["timeout"]
+        if kwargs.get("max_age"):
+            payload["maxAge"] = kwargs["max_age"]
+
+        result = await _firecrawl_scrape(provider_settings, payload)
+        content = result.get(output_format, "")
+        result_url = result.get("url") or url
+        ret = f"URL: {result_url}\nContent: {content}" if content else ""
+        return ret or "Error: Firecrawl web scraper does not return any results."
+
+
 @builtin_tool(config=_BAIDU_WEB_SEARCH_TOOL_CONFIG)
 @pydantic_dataclass
 class BaiduWebSearchTool(FunctionTool[AstrAgentContext]):

dashboard/src/components/chat/MessageListDEPRECATED.vue

@@ -303,7 +303,7 @@ export default {
                     part.tool_calls.forEach(toolCall => {
                         // 检查是否是支持引用解析的 web_search 工具调用
                         if (
-                            !['web_search_baidu', 'web_search_tavily', 'web_search_bocha', 'web_search_brave'].includes(toolCall.name) ||
+                            !['web_search_baidu', 'web_search_tavily', 'web_search_bocha', 'web_search_brave', 'web_search_firecrawl'].includes(toolCall.name) ||
                             !toolCall.result
                         ) {
                             return;

dashboard/src/i18n/locales/en-US/features/config-metadata.json

@@ -125,6 +125,10 @@
           "description": "Brave Search API Key",
           "hint": "Multiple keys can be added for rotation."
         },
+        "websearch_firecrawl_key": {
+          "description": "Firecrawl API Key",
+          "hint": "Multiple keys can be added for rotation."
+        },
         "websearch_baidu_app_builder_key": {
           "description": "Baidu Qianfan Smart Cloud APP Builder API Key",
           "hint": "Reference: [https://console.bce.baidu.com/iam/#/iam/apikey/list](https://console.bce.baidu.com/iam/#/iam/apikey/list)"

dashboard/src/i18n/locales/ru-RU/features/config-metadata.json

@@ -125,6 +125,10 @@
                     "description": "API-ключ Brave Search",
                     "hint": "Можно добавить несколько ключей для ротации."
                 },
+                "websearch_firecrawl_key": {
+                    "description": "API-ключ Firecrawl",
+                    "hint": "Можно добавить несколько ключей для ротации."
+                },
                 "websearch_baidu_app_builder_key": {
                     "description": "API-ключ Baidu Qianfan APP Builder",
                     "hint": "Ссылка: [https://console.bce.baidu.com/iam/#/iam/apikey/list](https://console.bce.baidu.com/iam/#/iam/apikey/list)"

dashboard/src/i18n/locales/zh-CN/features/config-metadata.json

@@ -127,6 +127,10 @@
           "description": "Brave Search API Key",
           "hint": "可添加多个 Key 进行轮询。"
         },
+        "websearch_firecrawl_key": {
+          "description": "Firecrawl API Key",
+          "hint": "可添加多个 Key 进行轮询。"
+        },
         "websearch_baidu_app_builder_key": {
           "description": "百度千帆智能云 APP Builder API Key",
           "hint": "参考:[https://console.bce.baidu.com/iam/#/iam/apikey/list](https://console.bce.baidu.com/iam/#/iam/apikey/list)"