@conxt/mcp
@conxt/mcp
面向本地优先的受治理 AI 记忆 MCP 服务器。 你的智能体提议记忆;由你批准;只有已批准的记忆才会被召回。无需账户即可独立运行——可选同步到 Conxt 云端,实现团队共享记忆。
实现开放的 Governed Memory Record 格式。
快速开始
让你的 IDE 直接指向它——无需安装,无需注册:
Cursor(~/.cursor/mcp.json)
{ "mcpServers": { "conxt": { "command": "npx", "args": ["-y", "@conxt/mcp"] } } }Claude Code
claude mcp add conxt -- npx -y @conxt/mcpClaude Desktop → 设置 → 开发者 → 编辑配置:与 Cursor 相同的配置块。
就这么简单。记忆存储在你机器上的 ~/.conxt/memory.json 中。
Related MCP server: Centralaizer
工具
工具 | 谁调用它 | 功能 |
| 智能体 | 提议一条记忆(待批准——尚未生效) |
| 智能体 | 针对查询召回已批准的记忆 |
| 智能体/你 | 查看等待批准的内容 |
| 你 | 批准一条待批准记忆 → 它现在生效 |
| 你 | 拒绝或删除 |
使其成为受治理的规则:智能体可以提议,但只有人类才能批准——并且 recall 永远不会返回任何未经批准的内容。
从终端进行人工批准
npx @conxt/mcp pending # what's waiting
npx @conxt/mcp approve <id> # approve it
npx @conxt/mcp reject <id>
npx @conxt/mcp list
npx @conxt/mcp path # where the store lives单人使用想要零摩擦?设置 CONXT_AUTO_APPROVE=1,你 remember 的所有内容都会立即被批准。
云端模式(可选)
本地始终是事实来源。同时设置两个环境变量,即可将已批准的记录推送到你的 Conxt 工作区,实现共享的、受治理的团队记忆:
CONXT_API_KEY=... CONXT_API_URL=https://...同步失败绝不会影响你的本地记忆。
配置
环境变量 | 默认值 | 用途 |
|
| 本地存储路径 |
| 关闭 | 在 |
| 未设置 | 可选的云端同步 |
开发
npm install && npm run build && node dist/index.js许可证
MIT。
发布
CI(.github/workflows/ci.yml)在每次推送时构建并运行治理冒烟测试。要发布到 npm:
在 npmjs.com 上,确认你拥有
@conxt作用域——或者将package.json中的包重命名为无作用域的conxt-mcp。添加
NPM_TOKEN(自动化令牌)作为仓库密钥(设置 → 密钥和变量 → Actions)。发布一个版本:
npm version patch # or minor / major — creates a v* tag git push --follow-tagspublish工作流会构建、进行冒烟测试,并运行npm publish --access public。
或者手动发布:
npm install && npm run build && npm login && npm publish --access publicMaintenance
Resources
Unclaimed servers have limited discoverability.
Looking for Admin?
If you are the server author, to access and configure the admin panel.
Related MCP Servers
- AlicenseNot gradedqualityCmaintenanceA local-first MCP memory server providing persistent, searchable memory for AI agents, powered by SQLite.51Apache 2.0
- FlicenseNot gradedqualityCmaintenanceA local-first, privacy-preserving centralized memory hub for AI agents — MCP-compatible, zero cloud egress.3
- AlicenseNot gradedqualityCmaintenanceA local-first memory server for AI tools that provides cross-tool, consent-gated, and auditable memory management via MCP.AGPL 3.0
- AlicenseNot gradedqualityAmaintenanceA local-first MCP server that gives AI clients a shared, user-owned memory and skills plane, enabling persistent memory across apps via the Model Context Protocol.MIT
Related MCP Connectors
Person-owned, portable AI memory as a remote MCP server, readable and writable by any MCP client.
Cloud-hosted MCP server for durable AI memory
Private-by-default, local-first memory/context/task orchestrator for MCP apps and agents.
Latest Blog Posts
- Who's Calling? MCP Hosts Are an Identity Blind Spot (And the Spec Knows It)By Om-Shree-0709 on .mcpAgent IdentityOAuth 2.1
- Your AI Chatbot Just Exposed Your CEO's Salary to an InternBy Om-Shree-0709 on .Agent IdentityMCP SecurityOAuth Delegation
- Why MCP Servers Need Execution Sandboxing (And Why Your Current Stack Isn't Enough)By Om-Shree-0709 on .Agentic AiPrompt InjectionWebAssembly
MCP directory API
We provide all the information about MCP servers via our MCP API.
curl -X GET 'https://glama.ai/api/mcp/v1/servers/pgavali0318/conxt-mcp'
If you have feedback or need assistance with the MCP directory API, please join our Discord server