awesome-trading-agents — Trading Agents + MCP List

Main

• Treat this as a map, not a recommendation: trading is risky—evaluate each project’s safety and constraints before running money flows.

• Start with data-only MCPs and research agents; only consider execution connectors after you have monitoring and limits.

• Use the “Agents / MCPs / Skills” split to build a modular stack: one agent, one data MCP, one skill pack.

• Keep everything paper-trading first: log decisions, compare to baselines, and only then consider live execution.

README (excerpt)

English · 简体中文

Awesome Trading Agents collects open-source projects where LLMs help research markets, make trading decisions, or connect agents to market data and execution tools. The list focuses on three building blocks: Agents, MCPs, and Skills. It does not try to cover classic quant libraries, time-series models, or reinforcement-learning trading bots; those are better served by georgezouq/awesome-ai-in-finance and wilsonfreitas/awesome-quant. Entries are selected for public code or artifacts, clear LLM-driven behavior, recent activity, useful documentation, a distinct role, and visible adoption. Stewarded by the LLMQuant community.

[!TIP] If you only read three:

• Agents — TauricResearch/TradingAgents · virattt/ai-hedge-fund · HKUDS/AI-Trader
• MCPs — alpacahq/alpaca-mcp-server · krakenfx/kraken-cli · financial-datasets/mcp-server

Source-backed notes

• README explains the list focuses on three building blocks: Agents, MCPs, and Skills, and includes a “If you only read three” starter section.
• README states the repo is bilingual (English + Simplified Chinese) and links a Chinese README.
• GitHub metadata verifies CC0-1.0 license, stars, and last push date for attribution.

FAQ

• Is this financial advice?: No—it's a curated list of open-source projects; use at your own risk and follow local laws.
• What should I read first?: The “If you only read three” section, then pick one agent + one data MCP to test.
• Can I use it with non-trading tasks?: Yes—many components (data MCPs, skills) can be reused for research workflows.