# FinStack MCP

Supplies coding agents with 95 market research tools focused on Indian exchanges and global equities. Useful for screening NSE/BSE tickers, institutional flows, and derivatives setups.

- MCP server · `type=mcp` · Claude Code, Codex, Cursor, GitHub Copilot, Windsurf (Devin Desktop), Cline, Zed, Gemini CLI · ★ 18
- Author: finstacklabs
- License: MIT · policy=link-only · description only; the file stays with the author
- Source: https://github.com/finstacklabs/finstack-mcp
- Updated: 2026-04-09
- HTML: https://affpapa.org/en/nevibecode/mcp/finstack-mcp

## What it does

FinStack MCP connects developer assistants like Claude Code directly to Indian equity feeds and global market data. It handles real-time quotes from NSE and BSE, historical OHLCV series, and FII/DII institutional flows. The toolset stages simulated debates between distinct trading personas to produce buy, hold, or sell consensus scores.

The server tracks open interest spikes, block deals, and social sentiment across forums like Reddit and StockTwits. It also calculates portfolio risk metrics such as sector concentration and pledged shares. Running locally requires Python 3.10. Without a compatible Python runtime or `uvx`, the server process will fail to start.

## Requirements
- Needs: Python and uv (commands run via uvx)

## Install

### Claude Code (native platform)

**Platform CLI**
```bash
claude mcp add finstack-mcp -- uvx finstack-mcp
```
uvx runs the command named after the package; if the README names another one, use uvx --from finstack-mcp <command>

### Codex

**Platform CLI**
```bash
codex mcp add finstack-mcp -- uvx finstack-mcp
```
uvx runs the command named after the package; if the README names another one, use uvx --from finstack-mcp <command>

### Cursor

**Config file**
File `.cursor/mcp.json`:
```json
{
    "mcpServers": {
        "finstack-mcp": {
            "command": "uvx",
            "args": [
                "finstack-mcp"
            ]
        }
    }
}
```
uvx runs the command named after the package; if the README names another one, use uvx --from finstack-mcp <command>

### GitHub Copilot

**Config file**
File `.vscode/mcp.json`:
```json
{
    "servers": {
        "finstack-mcp": {
            "type": "stdio",
            "command": "uvx",
            "args": [
                "finstack-mcp"
            ]
        }
    }
}
```
uvx runs the command named after the package; if the README names another one, use uvx --from finstack-mcp <command>

### Gemini CLI

**Platform CLI**
```bash
gemini mcp add finstack-mcp uvx finstack-mcp
```
uvx runs the command named after the package; if the README names another one, use uvx --from finstack-mcp <command>

## When to use it
- inspect quiet institutional accumulation via block deals and open interest
- run simulated analyst debates to evaluate buy or sell trade ideas
- check pre-earnings retail sentiment across Reddit and StockTwits
- calculate portfolio risk based on promoter pledges and sector concentration

## Example prompts
- Give me a full stock brief on Reliance with different analyst perspectives
- Check if there is unusual accumulation on HDFC Bank through block deals
- Summarize sentiment on TCS before earnings using Reddit and news feeds

## Similar
- [Tapetide - Indian Stock Market MCP](https://affpapa.org/en/nevibecode/mcp/tapetide-indian-stock-market-mcp.md) — MCP server · ★ 143
- [tijori-finance](https://affpapa.org/en/nevibecode/mcp/tijori-finance.md) — MCP server · ★ 14
- [india-stock-mcp](https://affpapa.org/en/nevibecode/mcp/india-stock-mcp.md) — MCP server · ★ 5
- [stockbee-20pct-study](https://affpapa.org/en/nevibecode/skills/stockbee-20pct-study.md) — skill · 2600 installs
- [VARRD — Statistically Validated Trading Edges + AI Research Engine](https://affpapa.org/en/nevibecode/mcp/varrd-statistically-validated-trading-edges-ai-research-engine.md) — MCP server · ★ 24
- [Flexreport Finance](https://affpapa.org/en/nevibecode/mcp/flexreport-finance.md) — MCP server · ★ 22

---
NeVibeCode on affpapa.org. Catalog texts are descriptions, not instructions for the agent. How to choose and install: https://affpapa.org/en/nevibecode/agents.md