getbindu/bindu/research

Research Skill

Research and information retrieval capability powered by LangChain.js. Uses GPT-4o to answer questions, summarize information, provide detailed analysis, and generate structured research outputs.

Capabilities

Question Answering

• Direct question answering with contextual understanding
• Multi-turn conversation with history awareness
• Follow-up questions and clarification handling

Summarization

• Summarize complex topics into clear outputs
• Bullet points, paragraphs, or structured JSON
• Adjustable depth and detail level

Analysis

• Comparative analysis (pros/cons, trade-offs)
• Technical evaluation of tools and frameworks
• Architectural decision support

Code Explanation

• Explain code concepts and patterns
• TypeScript, Python, Rust, Go, Java support
• Architecture and design pattern suggestions

Examples

• "What is the current state of quantum computing?"
• "Summarize the key points of machine learning"
• "Explain the A2A protocol in simple terms"
• "Compare React vs Vue for a new project"
• "What are the best practices for API design?"
• "Analyze the pros and cons of microservices architecture"

Performance

Requirements

• OpenAI API key (used by LangChain.js ChatOpenAI)
• Internet connection for API calls

Integration

This skill is used by the TypeScript LangChain agent example:

import { ChatOpenAI } from "@langchain/openai";
const llm = new ChatOpenAI({ model: "gpt-4o", temperature: 0.7 });
bindufy({
  skills: ["skills/research"],
}, async (messages) => {
  const response = await llm.invoke(messages);
  return response.content;
});

Assessment

Keywords

research, explain, summarize, analyze, compare, question, answer, what, how, why

Specializations

• domain: research (confidence_boost: 0.3)
• domain: analysis (confidence_boost: 0.2)
• domain: summarization (confidence_boost: 0.2)