Add guy-hartstein/company-research-agent

agents/guy-hartstein__company-research-agent/README.md

@@ -0,0 +1,57 @@
+# Agentic Company Researcher
+
+A production-ready, multi-agent research platform built on **LangGraph** and **Tavily** that generates comprehensive company research reports on demand.
+
+## What it does
+
+Given a company name, URL, headquarters, and industry, the agent orchestrates a full research pipeline:
+
+1. **Grounding** — Anchors context for the company before dispatching agents
+2. **Parallel Research** (four simultaneous specialists):
+   - **CompanyAnalyzer** — Core products, leadership team, business model, target market
+   - **IndustryAnalyzer** — Market size, growth, direct competitors, positioning
+   - **FinancialAnalyst** — Funding rounds, investors, revenue model, key metrics
+   - **NewsScanner** — Announcements, partnerships, press coverage, awards
+3. **Collector & Curator** — Aggregates, scores (Tavily relevance ≥ 0.4), and deduplicates research documents
+4. **Enricher** — Supplements curated data with additional context
+5. **Briefing** (Gemini 2.5 Flash) — Generates structured category briefings from high-context synthesis
+6. **Editor** (GPT-5.1) — Compiles all briefings into a polished, deduplicated markdown report with MLA references
+
+## Output format
+
+Reports follow a fixed schema:
+- **Company Overview** — Business model, leadership, products, differentiators
+- **Industry Overview** — Market position, competitors, challenges
+- **Financial Overview** — Funding history, investors, revenue model
+- **News** — Chronological events: announcements, partnerships, recognition
+- **References** — MLA-formatted citations
+
+## Key features
+
+- Dual-model architecture: Gemini 2.5 Flash for large-context synthesis + GPT-5.1 for precise formatting
+- Asynchronous processing via FastAPI with polling-based progress tracking
+- React frontend with real-time progress display and PDF export
+- Deployable via Docker or the included setup script
+- Live demo: https://companyresearcher.tavily.com
+
+## Example usage
+
+```bash
+# Start the backend
+python application.py
+
+# POST a research request
+curl -X POST http://localhost:8000/research \
+  -H "Content-Type: application/json" \
+  -d '{"company": "Stripe", "url": "https://stripe.com", "hq_location": "San Francisco, CA", "industry": "FinTech"}'
+```
+
+## Setup
+
+Requires: `TAVILY_API_KEY`, `GEMINI_API_KEY`, `OPENAI_API_KEY`
+
+```bash
+git clone https://github.com/guy-hartstein/company-research-agent.git
+cd company-research-agent
+./setup.sh   # auto-detects uv or pip
+```

agents/guy-hartstein__company-research-agent/metadata.json

@@ -0,0 +1,15 @@
+{
+  "name": "company-research-agent",
+  "author": "guy-hartstein",
+  "description": "Multi-agent pipeline using LangGraph and Tavily to generate deep company research reports covering business, industry, financials, and news.",
+  "repository": "https://github.com/guy-hartstein/company-research-agent",
+  "path": "",
+  "version": "1.0.0",
+  "category": "research",
+  "tags": ["langgraph", "tavily", "company-research", "multi-agent", "langchain", "gemini", "openai", "report-generation", "fastapi", "react"],
+  "license": "Apache-2.0",
+  "model": "google:gemini-2.5-flash",
+  "adapters": ["system-prompt"],
+  "icon": false,
+  "banner": false
+}