Log inSubmit server
Finlab AI logo

Finlab AI

Researches Taiwan-stock strategies with FinLab data, evidence, documentation, and backtesting examples through MCP.

InstallRepositoryWebsite
Full README Compare alternatives Embed widget Claim this listing

Install Finlab AI

Pick your client, then add it in one click or paste the config.

SSE (Remote)
Server URL
https://mcp.finlab.finance/mcp

Remote servers are added in the app, not the config file: Customize → Connectors → Add custom connector, then paste this URL.

Overview

The finlab-ai MCP server exposes FinLab’s quantitative trading resources through a hosted, read-only streamable HTTP endpoint. Its tools list strategy examples, return strategy code, retrieve stock evidence, browse a catalog of 900+ data columns, and access FinLab documentation. A shell installer can also set up the local skill for Claude Code, Codex, or Gemini. Reach for it when an agent needs FinLab-specific research references before implementing or evaluating a Taiwan-stock strategy.

Use cases

  • Discover quantitative strategy examples
  • Inspect FinLab Python implementations
  • Browse stock data columns
  • Retrieve evidence for stock research
  • Review backtesting documentation

Key features

  • Hosted read-only streamable HTTP MCP server
  • 900+ data columns across 80+ tables
  • 60+ complete strategy examples
  • Strategy retrieval with Python code
  • Stock evidence and data catalog tools
  • Backtesting and machine-learning references

Tools6

What Finlab AI lets your agent do. Expand a tool for its parameters.

Verified live~456 tokens of context

Security & health

What it can reach and how reliably it runs — from our own checks, not the maintainer’s claims.

ActiveLast checked Today

Recent health check succeeded.

7/7 checks healthy over the last 34d

What it can do

From the 6 tools this server returned when we ran it.

Read-only tools · 6

Observed 2026-10-15 in an isolated sandbox. A description of reach, not a security audit.

Adoption & maintenance

Factual signals from GitHub, npm, and our automated checks — not a rating.

GitHub stars416Stargazers on the source repository.
Last commit1d agoMost recent push to the default branch.
Tools exposed6Callable tools this server registers over MCP.
Directory activity3 viewsConfig copies, upvotes, and views on AllMCPs.

How Finlab AI works

What finlab-ai MCP server does

The finlab-ai MCP server gives AI clients read-only access to FinLab research material for quantitative stock analysis. It is an official FinLab product and focuses on strategy discovery, data references, stock evidence, and practical documentation rather than direct trading execution.

The available reference material covers more than 900 data columns across over 80 tables. It also includes the sim() backtesting API, resampling and performance metrics, more than 60 complete factor or strategy examples, machine-learning guidance for feature engineering and labels, and recommendations about common research patterns and anti-patterns.

How it works

The hosted service uses the MCP streamable HTTP transport. No local package installation is required to use that endpoint. Claude Code can register it with the finlab name and the URL https://mcp.finlab.finance/mcp.

The finlab-ai MCP server is read-only. The former finlab-ai.koreal6803.workers.dev address is retired; clients should use the current MCP URL.

A separate shell installer configures a local skill. The installer detects Claude Code, Codex, or Gemini, installs uv when necessary, and completes the skill setup. This provides an alternative to connecting directly to the hosted MCP endpoint.

Tools and capabilities

The server exposes these tools:

  • list_strategies for discovering available strategy examples.
  • get_strategy for retrieving a selected strategy, including Python code.
  • get_stock_evidence for obtaining stock-related evidence.
  • get_data_catalog for exploring the available FinLab data fields.
  • get_finlab_docs for accessing the skill documentation kept in sync with the main project.
  • how_to_start for orientation and initial usage guidance.

These capabilities suit research workflows that need to identify candidate factors, inspect example implementations, locate relevant data fields, or understand FinLab APIs before writing analysis code. The reference set describes Taiwan-stock strategy backtesting and includes material on single-factor and multi-factor approaches.

Setup and configuration

For the local skill, run:

Terminal
curl -sSf https://ai.finlab.finance/install.sh | sh

The installer handles CLI detection and installs uv if it is missing. For the hosted service, add the streamable HTTP endpoint to an MCP-compatible client. The README provides this Claude Code command:

Terminal
claude mcp add --transport http finlab https://mcp.finlab.finance/mcp

Read the full README View source on GitHub

Reviews

No reviews yet — be the first to share how it worked for you.

Frequently Asked Questions about Finlab AI

finlab-ai is an MCP server that connects AI clients to FinLab’s quantitative trading resources through a read-only streamable HTTP endpoint. Its main tools list strategies, retrieve strategy Python code, provide stock evidence, browse the data catalog, expose FinLab documentation, and explain how to get started.

Run `curl -sSf https://ai.finlab.finance/install.sh | sh` to install the local skill. The installer detects Claude Code, Codex, or Gemini, installs `uv` if needed, and configures the skill. You can instead connect directly to the hosted MCP endpoint without a local installation.

Run `claude mcp add --transport http finlab https://mcp.finlab.finance/mcp` in Claude Code. This registers the hosted, read-only streamable HTTP server under the name `finlab`, using the current endpoint rather than the retired Workers URL.

finlab-ai provides `list_strategies`, `get_strategy`, `get_stock_evidence`, `get_data_catalog`, `get_finlab_docs`, and `how_to_start`. Together, they support strategy discovery, code retrieval, stock evidence lookup, data-field exploration, documentation access, and initial orientation.

Related MCP servers

All in Finance & Fintech All alternatives
Add a badge to your READMEShow it’s listed — live SVG, Markdown or HTML.

AllMCPs Directory Badge

Full Badge Customizer

Showcase your server listing on GitHub or your project documentation. Embed this dynamic SVG badge to highlight official listing status and live engagement.

Badge Style:
Live Dynamic SVG PreviewFinlab AI AllMCPs Directory Badge
Markdown (GitHub README)
[![AllMCPs](https://allmcps.com/api/badge/finlab-ai?style=directory)](https://allmcps.com/mcp/finlab-ai)
HTML Embed
<a href="https://allmcps.com/mcp/finlab-ai"><img src="https://allmcps.com/api/badge/finlab-ai?style=directory" alt="Finlab AI on AllMCPs" /></a>