MCP.Market

LingJing MCP server

by genesis-plan·io.github.genesis-plan/lingjing·v1.0.2

LingJing classroom MCP: human teaches, AI student questions; returns blindspots & summary.

A89/100grade A
What users say
No reviews yet
Be the first
Safety scan
A89/100

not scanned yet

Adoption
New

0 stars

Reviews

Write one

Nobody has reviewed LingJing yet.

If you have run it, two minutes of your experience saves the next person an afternoon.

LingJing tools

No tool declarations could be read from the package source. They show once the server is installed.

Not scanned yet. New listings are scanned within a day or two of arriving.

How LingJing compares

Against every graded MCP server, from our own scans and uptime checks. Updated after each scan run.

Safety score
89/100
higher than 80% of 48,255 servers

Release history: 2 versions in the registry since 2026-10-09, latest 2026-10-13, typically one every 4 days.

Install LingJing in Claude Code, Cursor or VS Code

Runs npx -y lingjing-classroom-mcp on your machine. Read the scan report first; the gateway never runs local packages.

claude mcp add lingjing -- npx -y lingjing-classroom-mcp
Add to Cursor

What the publisher says

From the LingJing repository's README, as published. We do not edit it. Read it on GitHub

灵境 LingJing

一间虚拟课室:人类给 AI 上课,AI 当一面镜子。

人讲 → AI 多角追问 → 人答 → 人自判 → 带走你的思考、盲区与总结。

English web version (no install, BYO-key): — three personal uses: self-learner comprehension checks, a private knowledge library, your own experience made queryable. No signup; your API key stays in your browser and is never sent to any server (BYO-key app). Works with free models (OpenRouter free tier).

角色契约:人类是体验者(也是传授者),AI 是激发者;AI 不学习、不评分、不判定,判定权永远归人。 这不是产品态度,是架构不变量——世界模型 World = ⟨S, R, M, T⟩ 里根本不存在「学生 → 分数」这条边。

30 秒上手

① 跑起来(零依赖,Node 22+)

git clone https://github.com/genesis-plan/lingjing.git
cd lingjing
node server.js          # → http://localhost:8080

浏览器打开 (2D 课室)或 (3D 课室,Three.js 已本地化,离线可用)。 再点 / 上的输入框,讲一段你熟悉的东西——课就开始。

不配任何密钥也能跑:走确定性兜底语料,流程完整、体验降级、全程零外发。

② 想让 AI 学生说人话(可选,须自带 key)

# 默认硅基流动(国内,数据不出境);须自带 key
LINGJING_LLM_PROVIDER=siliconflow LINGJING_SILICONFLOW_KEY=xxx node server.js

# 或换其他境内提供商
LINGJING_LLM_PROVIDER=zhipu LINGJING_ZHIPU_KEY=xxx node server.js

# 仅当你明确要用境外 OpenRouter 时才指定(课堂文本将出境,须自担合规)
LINGJING_LLM_PROVIDER=openrouter LINGJING_OR_KEY=sk-or-v1-xxx node server.js

# 或换境内提供商(数据不出境)
LINGJING_LLM_PROVIDER=zhipu       LINGJING_ZHIPU_KEY=xxx       node server.js
LINGJING_LLM_PROVIDER=siliconflow LINGJING_SILICONFLOW_KEY=xxx node server.js
LINGJING_LLM_PROVIDER=deepseek    LINGJING_DEEPSEEK_KEY=xxx    node server.js

③ 看真实测试基线

node tools/test_questioning.mjs     # 提问引擎(确定性提问规格)
node tools/test_reflection.mjs      # 双稿制反思引擎
node tools/test_weakpoint.mjs       # 薄弱点定位层(跨场景必不同)
node tools/test_p0_wiring.mjs       # 盲区环端到端接线

完整 23 个测试的分类、依赖与真实通过数见 02 · 使用指南。

能力边界(诚实声明)

不承诺:AI 学生问出的话一定切中要害;任何"教育效果"。研究引用是我们的设计依据,不是"已达到该效果"的证据。

灵境 MCP · 给 AI Agent 的课室

灵境除了本机网页课室,现已打包成 MCP stdio server(lingjing-classroom-mcp)——让任何支持 MCP 的 AI Agent / 编排器直接调用这间"课室":人(或 Agent)讲授,灵境当学生追问,返回盲区与总结。

特性:确定性引擎,无需 LLM、无需联网、可离线复现;fail-closed(无效会话直接报错)。与网页课室共享同一套"七类认知盲区"判定逻辑,但不经过 LLM 嗓音层,纯本地可复现。

一键接入(agents npx 即用)

npx lingjing-classroom-mcp

四个工具:lingjingstartsession(开课)→ lingjingteach(讲授 + 苏格拉底追问)→ lingjingreflect(七维盲区 + 总结 + 卡片)→ lingjingclosesession(收尾并清理)。完整用法见 mcp/README.md。

发现 / 分发

Shortened. The full README is on GitHub.

Nothing above is checked by us. What we check is on the safety report.

LingJing: common questions

Is LingJing MCP server safe?
Yes, by our scan: it is graded A (89/100). Read the LingJing safety report
How do I install LingJing?
It runs on your machine. Copy the Claude Code, Cursor, VS Code or Claude Desktop config from the install section.
Does LingJing need an API key?
Not as far as the registry entry and our scan can tell: no credentials are declared or required.
Is LingJing maintained?
The last commit was 26 days ago (2026-10-09). The latest release is v1.0.2.

More from genesis-plan →