Merge pull request 'Add SearXNG web search for current information' (#8) from feature/internet-access into main

Reviewed-on: MSC/Daemon-Boyfriend#8

.env.example

@@ -49,6 +49,18 @@ BOT_STATUS=for mentions
 # Number of messages to remember per user (higher = more context, more tokens)
 MAX_CONVERSATION_HISTORY=20
 
+# ===========================================
+# Web Search (SearXNG)
+# ===========================================
+# SearXNG instance URL for web search (enables the bot to access current information)
+SEARXNG_URL=https://search.example.com
+
+# Enable/disable web search capability (true/false)
+SEARXNG_ENABLED=true
+
+# Maximum number of search results to fetch (1-20)
+SEARXNG_MAX_RESULTS=5
+
 # ===========================================
 # Logging
 # ===========================================

CLAUDE.md

@@ -39,11 +39,21 @@ Cogs are auto-loaded by `bot.py` from the `cogs/` directory.
 ### Configuration
 All config flows through `config.py` using pydantic-settings. The `settings` singleton is created at module load, so env vars must be set before importing.
 
+### Web Search
+The bot can search the web for current information via SearXNG:
+- `services/searxng.py` provides `SearXNGService` for web queries
+- `ai_chat.py` uses a two-step approach: first asks AI if search is needed, then provides results as context
+- Search is triggered automatically when the AI determines the query needs current information
+- Configured via `SEARXNG_URL`, `SEARXNG_ENABLED`, and `SEARXNG_MAX_RESULTS` env vars
+
 ### Key Design Decisions
 - `ConversationManager` stores per-user chat history in memory with configurable max length
 - Long AI responses are split via `split_message()` in `ai_chat.py` to respect Discord's 2000 char limit
 - The bot responds only to @mentions via `on_message` listener
+- Web search uses AI to decide when to search, avoiding unnecessary API calls for general knowledge questions
 
 ## Environment Variables
 
 Required: `DISCORD_TOKEN`, plus one of `OPENAI_API_KEY`, `OPENROUTER_API_KEY`, `ANTHROPIC_API_KEY`, or `GEMINI_API_KEY` depending on `AI_PROVIDER` setting.
+
+Optional: `SEARXNG_URL` for web search capability.

README.md

@@ -5,6 +5,7 @@ A customizable Discord bot that responds to @mentions with AI-generated response
 ## Features
 
 - **Multi-Provider AI**: Supports OpenAI, OpenRouter, Anthropic (Claude), and Google Gemini
+- **Web Search**: Access current information via SearXNG integration
 - **Fully Customizable**: Configure bot name, personality, and behavior
 - **Conversation Memory**: Remembers context per user
 - **Easy Deployment**: Docker support included
@@ -75,6 +76,16 @@ All configuration is done via environment variables in `.env`.
 | `AI_TEMPERATURE` | `0.7` | Response creativity (0.0-2.0) |
 | `MAX_CONVERSATION_HISTORY` | `20` | Messages to remember per user |
 
+### Web Search (SearXNG)
+
+| Variable | Default | Description |
+|----------|---------|-------------|
+| `SEARXNG_URL` | (none) | SearXNG instance URL |
+| `SEARXNG_ENABLED` | `true` | Enable/disable web search |
+| `SEARXNG_MAX_RESULTS` | `5` | Max search results to fetch |
+
+When configured, the bot automatically searches the web for queries that need current information (news, weather, etc.).
+
 ### Example Configurations
 
 **Friendly Assistant:**
@@ -143,7 +154,8 @@ src/daemon_boyfriend/
 └── services/
     ├── ai_service.py   # AI provider factory
     ├── providers/      # AI providers
-    └── conversation.py # Chat history
+    ├── conversation.py # Chat history
+    └── searxng.py      # Web search service
 ```
 
 ## License

src/daemon_boyfriend/cogs/ai_chat.py

@@ -7,7 +7,7 @@ import discord
 from discord.ext import commands
 
 from daemon_boyfriend.config import settings
-from daemon_boyfriend.services import AIService, ConversationManager, Message
+from daemon_boyfriend.services import AIService, ConversationManager, Message, SearXNGService
 
 logger = logging.getLogger(__name__)
 
@@ -71,6 +71,9 @@ class AIChatCog(commands.Cog):
         self.bot = bot
         self.ai_service = AIService()
         self.conversations = ConversationManager()
+        self.search_service: SearXNGService | None = None
+        if settings.searxng_enabled and settings.searxng_url:
+            self.search_service = SearXNGService(settings.searxng_url)
 
     @commands.Cog.listener()
     async def on_message(self, message: discord.Message) -> None:
@@ -138,10 +141,23 @@ class AIChatCog(commands.Cog):
         # Add current message to history for the API call
         messages = history + [Message(role="user", content=user_message)]
 
+        # Check if we should search the web
+        search_context = await self._maybe_search(user_message)
+
+        # Build system prompt with search context if available
+        system_prompt = self.ai_service.get_system_prompt()
+        if search_context:
+            system_prompt += (
+                "\n\n--- Web Search Results ---\n"
+                "Use the following current information from the web to help answer the user's question. "
+                "Cite sources when relevant.\n\n"
+                f"{search_context}"
+            )
+
         # Generate response
         response = await self.ai_service.chat(
             messages=messages,
-            system_prompt=self.ai_service.get_system_prompt(),
+            system_prompt=system_prompt,
         )
 
         # Save the exchange to history
@@ -154,6 +170,64 @@ class AIChatCog(commands.Cog):
 
         return response.content
 
+    async def _maybe_search(self, query: str) -> str | None:
+        """Determine if a search is needed and perform it.
+
+        Args:
+            query: The user's message
+
+        Returns:
+            Formatted search results or None if search not needed/available
+        """
+        if not self.search_service:
+            return None
+
+        # Ask the AI if this query needs current information
+        decision_prompt = (
+            "You are a search decision assistant. Your ONLY job is to decide if the user's "
+            "question requires current/real-time information from the internet.\n\n"
+            "Respond with ONLY 'SEARCH: <query>' if a web search would help answer the question "
+            "(replace <query> with optimal search terms), or 'NO_SEARCH' if the question can be "
+            "answered with general knowledge.\n\n"
+            "Examples that NEED search:\n"
+            "- Current events, news, recent happenings\n"
+            "- Current weather, stock prices, sports scores\n"
+            "- Latest version of software, current documentation\n"
+            "- Information about specific people, companies, or products that may have changed\n"
+            "- 'What time is it in Tokyo?' or any real-time data\n\n"
+            "Examples that DON'T need search:\n"
+            "- General knowledge, science, math, history\n"
+            "- Coding help, programming concepts\n"
+            "- Personal advice, opinions, creative writing\n"
+            "- Explanations of concepts or 'how does X work'"
+        )
+
+        try:
+            decision = await self.ai_service.chat(
+                messages=[Message(role="user", content=query)],
+                system_prompt=decision_prompt,
+            )
+
+            response_text = decision.content.strip()
+
+            if response_text.startswith("SEARCH:"):
+                search_query = response_text[7:].strip()
+                logger.info(f"AI decided to search for: {search_query}")
+
+                results = await self.search_service.search(
+                    query=search_query,
+                    max_results=settings.searxng_max_results,
+                )
+
+                if results:
+                    return self.search_service.format_results_for_context(results)
+
+            return None
+
+        except Exception as e:
+            logger.warning(f"Search decision/execution failed: {e}")
+            return None
+
 
 async def setup(bot: commands.Bot) -> None:
     """Load the AI Chat cog."""

src/daemon_boyfriend/config.py

@@ -61,6 +61,11 @@ class Settings(BaseSettings):
         20, description="Max messages to keep in conversation memory per user"
     )
 
+    # SearXNG Configuration
+    searxng_url: str | None = Field(None, description="SearXNG instance URL for web search")
+    searxng_enabled: bool = Field(True, description="Enable web search capability")
+    searxng_max_results: int = Field(5, ge=1, le=20, description="Maximum search results to fetch")
+
     def get_api_key(self) -> str:
         """Get the API key for the configured provider."""
         key_map = {

src/daemon_boyfriend/services/__init__.py

@@ -3,10 +3,12 @@
 from .ai_service import AIService
 from .conversation import ConversationManager
 from .providers import AIResponse, Message
+from .searxng import SearXNGService
 
 __all__ = [
     "AIService",
     "AIResponse",
     "Message",
     "ConversationManager",
+    "SearXNGService",
 ]

src/daemon_boyfriend/services/searxng.py

@@ -0,0 +1,107 @@
+"""SearXNG search service for web queries."""
+
+import logging
+from dataclasses import dataclass
+
+import aiohttp
+
+logger = logging.getLogger(__name__)
+
+
+@dataclass
+class SearchResult:
+    """A single search result."""
+
+    title: str
+    url: str
+    content: str
+
+
+class SearXNGService:
+    """Service for searching the web via SearXNG."""
+
+    def __init__(self, base_url: str, timeout: int = 10) -> None:
+        """Initialize the SearXNG service.
+
+        Args:
+            base_url: The base URL of the SearXNG instance
+            timeout: Request timeout in seconds
+        """
+        self.base_url = base_url.rstrip("/")
+        self.timeout = timeout
+
+    async def search(
+        self,
+        query: str,
+        categories: str = "general",
+        max_results: int = 5,
+    ) -> list[SearchResult]:
+        """Search the web using SearXNG.
+
+        Args:
+            query: The search query
+            categories: Search categories (general, images, news, etc.)
+            max_results: Maximum number of results to return
+
+        Returns:
+            List of search results
+        """
+        url = f"{self.base_url}/search"
+        params = {
+            "q": query,
+            "format": "json",
+            "categories": categories,
+        }
+
+        logger.debug(f"Searching SearXNG for: {query}")
+
+        try:
+            async with aiohttp.ClientSession() as session:
+                async with session.get(
+                    url,
+                    params=params,
+                    timeout=aiohttp.ClientTimeout(total=self.timeout),
+                ) as response:
+                    if response.status != 200:
+                        logger.error(f"SearXNG returned status {response.status}")
+                        return []
+
+                    data = await response.json()
+                    results = []
+
+                    for item in data.get("results", [])[:max_results]:
+                        results.append(
+                            SearchResult(
+                                title=item.get("title", ""),
+                                url=item.get("url", ""),
+                                content=item.get("content", ""),
+                            )
+                        )
+
+                    logger.debug(f"SearXNG returned {len(results)} results")
+                    return results
+
+        except aiohttp.ClientError as e:
+            logger.error(f"SearXNG request failed: {e}")
+            return []
+        except TimeoutError:
+            logger.error("SearXNG request timed out")
+            return []
+
+    def format_results_for_context(self, results: list[SearchResult]) -> str:
+        """Format search results as context for the AI.
+
+        Args:
+            results: List of search results
+
+        Returns:
+            Formatted string with search results
+        """
+        if not results:
+            return "No search results found."
+
+        formatted = []
+        for i, result in enumerate(results, 1):
+            formatted.append(f"[{i}] {result.title}\n    URL: {result.url}\n    {result.content}")
+
+        return "\n\n".join(formatted)