VCForge AI

AI venture due diligence agent team by Sharad Khare.

VCForge AI is a multi-agent startup investment analysis pipeline built with Google ADK and Gemini. It researches companies, models markets, evaluates risk, and generates investor-ready artifacts.

Built and maintained by Sharad Khare — AI strategist, full-stack developer, and builder of practical multi-agent systems.

---

What this project does

VCForge automates core due diligence workflows for venture and growth teams:

• company and founder research
• market sizing and competitive positioning
• financial scenario modeling
• multi-category risk scoring
• investor memo synthesis
• professional HTML report generation
• visual infographic summary

You provide a startup name, URL, or both. The pipeline produces structured analysis and downloadable outputs.

---

Why Sharad Khare built VCForge AI

Investment diligence is repetitive, research-heavy, and document-intensive. Teams need faster first-pass analysis without sacrificing structure.

VCForge demonstrates a production-style sequential agent architecture where each stage has a clear responsibility and output key, making the workflow auditable and extensible.

---

How it works

User Query (company/url + round context)
   │
   ▼
Stage 1: Company Research
   │
   ▼
Stage 2: Market Analysis
   │
   ▼
Stage 3: Financial Modeling (+ revenue chart)
   │
   ▼
Stage 4: Risk Assessment
   │
   ▼
Stage 5: Investor Memo
   │
   ▼
Stage 6: HTML Report
   │
   ▼
Stage 7: Infographic Summary

Pipeline stages

| Stage | Agent focus | Primary output |

|------|-------------|----------------|

| 1 | Company research | company_info |

| 2 | Market analysis | market_analysis |

| 3 | Financial modeling | financial_model + chart |

| 4 | Risk assessment | risk_assessment |

| 5 | Investor memo | investor_memo |

| 6 | Report generation | HTML investment report |

| 7 | Infographic generation | Visual TL;DR image |

---

Features

• Live web research with google_search
• URL-based startup analysis support
• Bear/Base/Bull revenue projection charts
• Five-category risk framework
• McKinsey-style HTML due diligence reports
• AI-generated infographic summaries
• ADK artifact versioning for generated files

---

Quick start

cd vcforge-ai
pip install -r requirements.txt
export GOOGLE_API_KEY=your_api_key
adk web

Then open the ADK UI and select the VCForge agent package.

Example prompts

• "Analyze https://example-startup.com for Series A investment of $30-50M"
• "Research Acme AI for its next funding round"
• "Evaluate Lovable for Series C funding opportunities"

---

Generated artifacts

Artifacts are saved in ADK and locally under outputs/:

• revenue_chart_*.png — financial projections
• investment_report_*.html — full diligence report
• infographic_*.png — visual executive summary

---

Project structure

vcforge-ai/
├── agent.py
├── tools.py
├── vcforge/
│   └── config.py
├── outputs/
├── requirements.txt
├── pyproject.toml
└── README.md

---

Who this is for

• Venture capital analysts and associates
• Angel investors evaluating startup opportunities
• Corporate venture teams running first-pass diligence
• Founders building investment intelligence copilots

Use cases

• Generate structured diligence memos from company name or URL
• Produce investor-ready HTML reports for internal review
• Model bear/base/bull scenarios for early investment decisions
• Create visual one-pagers for partner meetings

---

Author

• Website: sharadkhare.in

License

MIT © Sharad Khare