OpenGithubs / OpenGithubs/monthly
【开源自荐】QuantDinger:可自托管的 AI 量化交易工作台
Nobody has claimed this yet.
- Dominant language
- No language data
- Stars
- 851
- Forks
- 20
- PR merge metrics
- No merged PRs in 30d
Description
项目名称
QuantDinger
项目地址
https://github.com/OpenByteInc/QuantDinger
在线体验 / 文档
- 在线体验:https://ai.quantdinger.com
- 项目主页:https://github.com/OpenByteInc/QuantDinger
- Agent / MCP 文档:https://github.com/OpenByteInc/QuantDinger/tree/main/docs/agent
- MCP 源码与安装:https://github.com/OpenByteInc/QuantDinger/tree/main/mcp_server
- 策略开发指南:https://github.com/OpenByteInc/QuantDinger/blob/main/docs/trading/STRATEGY_DEV_GUIDE.md
项目简介
QuantDinger 是一个可自托管的开源 AI 量化交易工作台,把市场研究、Python 策略开发、历史回测、纸盘交易、实盘执行和监控放在一套流程中。后端采用 Python、Flask、Celery、PostgreSQL 和 Redis,策略代码、部署权限和经纪商凭据由操作者管理;项目同时提供 Agent Gateway 和 MCP 服务,方便 Claude Code、Cursor、Codex 等客户端调用受控的研究和回测工具。
推荐原因
- 完整闭环:从 AI 辅助研究、Strategy API V2 策略代码,到回测任务、纸盘和受控实盘执行。
- 可复核的工程边界:长时间策略运行由 trading worker 负责,有限回测和研究任务由 Celery 负责,持久状态进入 PostgreSQL,缓存与任务队列分离。
- MCP 与安全控制:MCP 默认 stdio,也支持 SSE 和 streamable HTTP;Agent Token 有作用域,变更操作需要幂等键,实盘权限默认关闭并需要额外确认。
- Python 友好:公开示例展示如何声明交易标的、频率、预热窗口,并通过目标仓位函数发出信号。
- 适合学习和二次开发:Apache-2.0 后端源码、Docker Compose 部署、API 与架构文档齐全。
快速开始
git clone https://github.com/OpenByteInc/QuantDinger.git
cd QuantDinger
docker compose up -d --build
策略流程可以从公开的 Strategy API V2 示例 开始,然后按照 MCP 文档 编译策略、提交回测并轮询任务结果。示例文件名使用了 EMA,但当前代码实际计算的是简单滚动均值;这里不把它包装成未经核实的收益案例。
适合分类
Python / AI Agent / 量化交易 / 回测 / 自托管 / MCP
License
Apache-2.0(后端);Web 和移动客户端仓库有各自的许可证与商业条款。
Contributor guide
No contributing guide indexed for this repository
First steps
- Read the whole issue, then the project's contributing guide.
- Comment on the issue to say you are picking it up — it saves two people doing the same work.
- Fork the repository and make your change on a branch.
- Open a pull request that references the issue number.
Research direction
The submission links QuantDinger's project homepage, Agent/MCP documentation, MCP source, and strategy development guide; read those materials first to verify the project's description. The issue does not name a file or specify the change required in the monthly repository, so completion would require an explicit publication or content decision.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- docker-compose, flask, postgresql, python, redis
- Domain
- content, documentation
- Issue type
- Documentation
- Difficulty
- 2/5
- Estimated time
- 1-3 hours
- Activity status
- Active
- Clarity
- Needs clarification
- Newbie friendliness
- 35/100