---
title: SEC10QInsight
emoji: ⚡
colorFrom: purple
colorTo: indigo
sdk: gradio
sdk_version: 5.33.0
app_file: app.py
pinned: false
license: mit
short_description: SEC 10-Q data explorer with AI insights
tags:
  - mcp-server-track
  - agent-demo-track
  - gradio-mcp-server
  - sec-10q
  - financial-insight

---

# `📘 SEC10QInsight – A Gradio-Based MCP Server for SEC Financial Analysis`

# 🔍 Overview – SEC10QInsight: Agentic SEC Data Analyst

👉 [**Jump to: Try It Yourself ⬇️**](#-try-it-yourself)

---

## 🎥 Demo Video

▶️ **Watch the full demo on YouTube:**  
[https://youtu.be/vin-Ovz2sFI?si=9n1kIcgnNs-b2K8d](https://youtu.be/vin-Ovz2sFI?si=9n1kIcgnNs-b2K8d)

[![Watch on YouTube](https://img.youtube.com/vi/vin-Ovz2sFI/0.jpg)](https://youtu.be/vin-Ovz2sFI?si=9n1kIcgnNs-b2K8d)


---

**SEC10QInsight** is an interactive MCP (Model Context Protocol) server built with **Gradio** that empowers users to query, visualize, and interpret financial data extracted from **SEC 10-Q filings**. Leveraging the power of large language models (LLMs), **SEC10QInsight** translates raw financial disclosures into insightful summaries, trend analysis, and visual dashboards.

Built with **Gradio** and powered by the **Model Context Protocol (MCP)**, this app simulates a multi-step agent process:
1. **Retrieve** real-time financial data from the SEC EDGAR database using a CIK (Central Index Key)
2. **Structure** and filter relevant metrics (e.g., `ComprehensiveIncomeNetOfTax`)
3. **Visualize** historical trends using line plots
4. **Analyze** the data using a large language model (LLM) to generate concise insights

---

## 🎯 Purpose

This app demonstrates how **agentic workflows** can be applied to **financial intelligence**, allowing a single natural language query to trigger a multi-step, automated reasoning process that includes:
- Data retrieval (from SEC)
- Contextual structuring and preprocessing
- Visualization
- LLM-driven interpretation

---

## 🔑 Key Features

- 🔍 **Natural language querying** of SEC data (e.g., "Show trends")
- 📊 **Interactive data visualization** of quarterly metrics like Comprehensive Income
- 🤖 **LLM-powered financial insights**, integrated with models like SambaNova
- 🌐 **Live data fetching** from the SEC's EDGAR API (XBRL format)
- ⚙️ **Gradio UI** for clean, user-friendly access to analysis tools

---

## 🛠️ Usage

1. **Select a company** (Apple, Tesla, Microsoft) (The sample here for HF MCP Hachathon specifically)
2. **Enter a query**, such as:
   - *"Summarize the trend"*
   - *"How did income change over time?"*
3. **Submit the query**
4. The app will:
   - Fetch SEC 10-Q data
   - Display it in a table
   - Render a trend plot
   - Return a model-generated interpretation

---

## 🤖 Agentic Highlights

- **Autonomous orchestration** of data + language processing
- **Modular architecture** with MCP for easy extensibility
- **Real-world grounding** using factual data from SEC filings

---

**SEC10QInsight** is ideal as a reference project or demo for:
- Agentic LLM workflows
- Financial data applications
- LLM × structured data integrations

Ideal for researchers, analysts, developers, and AI enthusiasts, **SEC10QInsight** demonstrates how open financial data and cutting-edge AI can work together to deliver explainable, real-time insights.

---

## 🚀 Try It Yourself

Users can try this amazing MCP server by accessing the client and code here:  
👉 [SEC10QInsight-MCP-CLIENT on Hugging Face Spaces](https://huggingface.co/spaces/Agents-MCP-Hackathon/SEC10QInsight-MCP-CLIENT)


---

## 📦 Tech Stack

- Python
- Gradio
- Pandas & Matplotlib
- SEC EDGAR API (XBRL JSON)
- OpenAI-compatible LLM client (e.g., SambaNova)
- MCP (Model Context Protocol)

---

## 📜 License

MIT License — open for learning, building, and extending.


Perfect for exploring how LLMs can assist in structured financial analysis through agentic design!