面向企业投研与产业研究团队的只读研究 MCP,提供研究框架、冻结快照与历史因子摘要,每条结论附来源、数据截至日与证伪条件,不输出交易建议。
git clone https://github.com/mikeli20221102-ux/diyan-research-mcpTools overview
# diyan-research-mcp · 企业研究证据链
<!-- mcp-name: io.github.mikeli20221102-ux/diyan-research-mcp -->
面向企业投研与产业研究团队的只读研究 MCP,提供研究框架、冻结快照与历史因子摘要,每条结论附来源、数据截至日与证伪条件,不输出交易建议。
## 这个仓库里没有数据
本仓库只包含转发壳,职责是三件事:校验入参、携带租户凭据调用研究数据服务、在返回前剥掉交易类字段。研究框架、冻结快照、假设台账和因子摘要都存放在服务端,不随代码分发。
```
AI 客户端 ──MCP──▶ diyan-research-mcp(本仓库) ──HTTPS──▶ 研究数据服务
入参校验 / 凭据转发 / 输出过滤 数据、租户、档位、配额、审计
```
这样分层有三个后果,都是有意为之:
- 壳可以公开审阅,因为它不含任何专有数据,也不含任何密钥。
- 租户身份、档位和配额一律由服务端裁定,壳不做信任判断,也就无法被绕过。
- 输出过滤留在壳内而不是只放在服务端,即便服务端出错或被篡改,交易类表述也无法离开进程。
## 工具
| 工具 | 说明 |
|---|---|
| `framework_excerpt` | 读取研究框架片段,可选 enterprise / business / finance / people / brand |
| `stock_snapshot_latest` | 读取最新冻结的股票研究快照 |
| `hypothesis_get` | 读取通过交付证据门槛的冻结假设 |
| `cognition_radar_read` | 读取历史主题研究雷达摘要 |
| `ic_backtest_summary` | 读取冻结的因子历史 IC 摘要 |
| `research_analyze_safe` | 调用受控研究接口做单标的研判 |
工具是否可用取决于凭据对应的档位。免费档开放研究框架与部分冻结快照,其余返回 `error_type: backend` 并说明需要付费租户凭据。
## 配置
唯一必填项是 `DIYAN_API_KEY`,在 MCP 详情页填写即可。其余变量见 `.env.example`,只在自建部署或联调时需要。
凭据缺失时进程仍会正常启动,但每次工具调用都会返回明确的配置错误。这是刻意的:托管环境可能在使用者填写凭据之前就拉起进程,启动即崩溃会让详情页上的服务看起来是坏的。
## 本地运行
```bash
python -m pip install -r requirements.txt
export PYTHONPATH=src
export DIYAN_API_KEY=<你的租户凭据>
export DIYAN_TRANSPORT=stdio
python -m diyan_research_mcp.server
```
测试不需要网络和凭据:
```bash
python -m unittest discover -s tests -v
```
## 边界
- 只读。没有任何写入、下单、部署或数据导出能力。
- 不返回实时行情。快照与摘要都是冻结产物,每条结果都带 `as_of`。
- 不生成买卖、仓位、股数、跟单或收益承诺。含此类表述的提问在发出请求前就会被拒绝。
- 不读取使用者的本地文件、本地数据库或非必要环境变量。
## 服务端契约
如需自建后端,接口定义见 [`docs/BACKEND_CONTRACT.md`](docs/BACKEND_CONTRACT.md)。
## 许可与归属
由深圳瞳桦文化传媒有限公司开发与维护。MIT 许可,见 [`LICENSE`](LICENSE)。
What people ask about diyan-research-mcp
What is mikeli20221102-ux/diyan-research-mcp?
+
mikeli20221102-ux/diyan-research-mcp is tools for the Claude AI ecosystem. 面向企业投研与产业研究团队的只读研究 MCP,提供研究框架、冻结快照与历史因子摘要,每条结论附来源、数据截至日与证伪条件,不输出交易建议。 It has 1 GitHub stars and was last updated today.
How do I install diyan-research-mcp?
+
You can install diyan-research-mcp by cloning the repository (https://github.com/mikeli20221102-ux/diyan-research-mcp) or following the README instructions on GitHub. ClaudeWave also provides quick install blocks on this page.
Is mikeli20221102-ux/diyan-research-mcp safe to use?
+
mikeli20221102-ux/diyan-research-mcp has not been audited yet by our security agent. Review the original repository on GitHub before using it in production.
Who maintains mikeli20221102-ux/diyan-research-mcp?
+
mikeli20221102-ux/diyan-research-mcp is maintained by mikeli20221102-ux. The last recorded GitHub activity is from today, with 0 open issues.
Are there alternatives to diyan-research-mcp?
+
Yes. On ClaudeWave you can browse similar tools at /categories/tools, sorted by popularity or recent activity.
Deploy diyan-research-mcp to your cloud
Ship this repo to production in minutes. Each platform spins up its own environment with editable env vars.
Maintain this repo? Add a badge to your README
Drop the badge into your GitHub README to show it's tracked on ClaudeWave. Each badge links back to this page and reflects the live Trust Score.
[](https://claudewave.com/repo/mikeli20221102-ux-diyan-research-mcp)<a href="https://claudewave.com/repo/mikeli20221102-ux-diyan-research-mcp"><img src="https://claudewave.com/api/badge/mikeli20221102-ux-diyan-research-mcp" alt="Featured on ClaudeWave: mikeli20221102-ux/diyan-research-mcp" width="320" height="64" /></a>More Tools
A single CLAUDE.md file to improve Claude Code behavior, derived from Andrej Karpathy's observations on LLM coding pitfalls.
An AI SKILL that provide design intelligence for building professional UI/UX multiple platforms
🪨 why use many token when few token do trick — Claude Code skill that cuts 65% of tokens by talking like caveman
CLI proxy that reduces LLM token consumption by 60-90% on common dev commands. Single Rust binary, zero dependencies
The fastest, litest AI Gateway. Rust core with Python SDK. Call 100+ LLM APIs in OpenAI (or native) format with cost tracking, guardrails, load balancing, and logging [Bedrock, Azure, OpenAI, Anthropic, OpenAI, VertexAI, vLLM, Nvidia NIM]
A collection of notebooks/recipes showcasing some fun and effective ways of using Claude.