Hiddenden/loyal_companion
3
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

Add SearXNG web search for current information

Browse Source 
- Add searxng.py service for web queries via SearXNG API
- Integrate search into ai_chat.py with AI-driven search decisions
- AI determines if query needs current info, then searches automatically
- Add SEARXNG_URL, SEARXNG_ENABLED, SEARXNG_MAX_RESULTS config options
- Update documentation in README.md, CLAUDE.md, and .env.example
This commit is contained in:
Latte
2026-01-11 20:49:20 +01:00
parent c5c42c8701
commit 6a9b6fdda2
7 changed files with 225 additions and 3 deletions
Download Patch File Download Diff File Expand all files Collapse all files

12
.env.example
View File

@@ -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
# ===========================================

10
CLAUDE.md
View File

@@ -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.

14
README.md
View File

@@ -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

78
src/daemon_boyfriend/cogs/ai_chat.py
View File

@@ -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."""

5
src/daemon_boyfriend/config.py
View File

@@ -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 = {

2
src/daemon_boyfriend/services/__init__.py
View File

@@ -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",
]

107
src/daemon_boyfriend/services/searxng.py Normal file
View File

@@ -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)