ResearchForge AI

Multi-agent research assistant by Sharad Khare.

ResearchForge AI uses OpenAI Agents SDK to plan research, gather web-backed insights, and generate structured long-form reports in Streamlit.

Features

• Multi-agent architecture: triage, research, and editor
• Web search-enabled evidence gathering
• Structured report output with outline, sources, and word count
• Fact capture helper tool with source attribution
• Download generated reports as markdown files

Quick start

cd researchforge-ai
pip install -r requirements.txt
streamlit run app.py

Required API key

• Set OPENAI_API_KEY in your environment before running the app.

How it works

1. You enter a topic in the sidebar.

2. The triage agent generates a focused research plan.

3. The research agent collects web findings and key facts.

4. The editor agent compiles a detailed markdown report.

Project structure

researchforge-ai/
├── app.py
├── researchforge/
│   ├── agents.py
│   ├── config.py
│   └── schemas.py
├── requirements.txt
├── pyproject.toml
└── README.md

Who this is for

• Analysts producing fast web-backed research briefs
• Founders validating ideas through structured AI research
• Content teams generating long-form reports from topics
• Developers learning OpenAI Agents SDK orchestration

Use cases

• Generate deep-dive reports from a single research prompt
• Coordinate triage, research, and editorial AI agents
• Capture cited facts and export markdown deliverables
• Build internal research copilots for knowledge teams

Author

• Website: sharadkhare.in

License

MIT © Sharad Khare