llamaindex-browseaidev 0.1.0

LlamaIndex integration for BrowseAI Dev — verified web search with citations and confidence scores for AI agents

llamaindex-browseaidev

LlamaIndex integration for BrowseAI Dev — verified web search with citations and confidence scores for AI agents.

Installation

pip install llamaindex-browseaidev

Quick Start

from llama_index.core.agent import ReActAgent
from llama_index.llms.openai import OpenAI
from llamaindex_browseaidev import BrowseAIDevAnswerTool, BrowseAIDevSearchTool

# Create tools
answer_tool = BrowseAIDevAnswerTool(api_key="bai_xxx")
search_tool = BrowseAIDevSearchTool(api_key="bai_xxx")

# Create agent
llm = OpenAI(model="gpt-4o")
agent = ReActAgent.from_tools([answer_tool, search_tool], llm=llm, verbose=True)

# Research with verified answers
response = agent.chat("What are the latest findings on mRNA vaccine efficacy?")
print(response)

Available Tools

• BrowseAIDevAnswerTool — Verified research with citations, confidence scores, contradiction detection
• BrowseAIDevSearchTool — Web search returning ranked results
• BrowseAIDevExtractTool — Extract structured claims from a URL
• BrowseAIDevCompareTool — Compare raw LLM vs verified answer

All tools are returned as FunctionTool instances, compatible with any LlamaIndex agent.

Why BrowseAI Dev?

Unlike raw search APIs, BrowseAI Dev fact-checks results. Every answer includes per-claim verification (BM25 + NLI), cross-source consensus, contradiction detection, and 7-factor evidence-based confidence scores. Open source (MIT).

Links

• Website · GitHub · Discord
• Python SDK · LangChain · CrewAI · MCP Server

License

MIT