Codex
БесплатноДавно не обновлялся (215д)Knowledge infrastructure SDK for AI agents - universal references, multi-tier caching, storage abstraction, and sync
Описание
Knowledge infrastructure SDK for AI agents - universal references, multi-tier caching, storage abstraction, and sync
README
Knowledge infrastructure for AI agents.
Overview
Fractary Codex provides infrastructure for managing organizational knowledge across AI agents and projects. It implements a universal reference system (codex:// URIs), multi-provider storage, intelligent caching, and file synchronization.
Packages
| Package | Description | Install |
|---|---|---|
| @fractary/codex | JavaScript/TypeScript SDK | npm install @fractary/codex |
| @fractary/codex-cli | Command-line interface | npm install -g @fractary/codex-cli |
| @fractary/codex-mcp | MCP server for AI agents | npx @fractary/codex-mcp |
| Claude Code Plugin | Claude Code integration | Install via plugin system |
Quick Start
New here? See the Getting Started Guide for a full walkthrough: creating a codex repository, setting up a GitHub token, and running your first sync.
1. Install
npm install -g @fractary/codex # SDK (required by CLI and MCP server)
npm install -g @fractary/codex-cli # CLI
2. Initialize configuration
Codex configuration lives in the codex: section of .fractary/config.yaml, alongside other Fractary plugin settings. This file is created by @fractary/core — initialize that first.
Recommended: Use the Claude Code plugin. The /fractary-codex-config init skill auto-detects your organization, project, and codex repository before writing anything.
Alternative: Use the CLI directly:
fractary-codex config-init --org myorg --codex-repo codex.myorg.com
3. Fetch documents
fractary-codex document-fetch codex://myorg/project/docs/api.md
4. Sync with codex repository
fractary-codex sync --dry-run
fractary-codex sync
5. Manage cache
fractary-codex cache-list --verbose
fractary-codex cache-stats
fractary-codex cache-clear --all
fractary-codex cache-health
6. Use programmatically
import { CodexClient } from '@fractary/codex'
// organizationSlug is optional — auto-detected from git remote if omitted
const client = await CodexClient.create({ organizationSlug: 'myorg' })
const result = await client.fetch('codex://myorg/project/docs/api.md')
console.log(result.content.toString())
MCP Server
Add to .mcp.json for AI agent integration:
{
"mcpServers": {
"fractary-codex": {
"command": "npx",
"args": ["-y", "@fractary/codex-mcp", "--config", ".fractary/config.yaml"]
}
}
}
Claude Code Plugin
The plugin provides skills for Claude Code, invocable as slash commands:
| Skill | Description |
|---|---|
/fractary-codex-config init |
Initialize codex configuration |
/fractary-codex-config update |
Update configuration fields |
/fractary-codex-config validate |
Validate configuration (read-only) |
/fractary-codex-sync |
Sync project with codex repository |
URI Format
codex://org/project/path/to/file.md
└─┬─┘└──┬──┘└──────┬───────┘
│ │ └─ File path within project
│ └─ Project/repository name
└─ Organization name
Project Structure
codex/
├── sdk/js/ # @fractary/codex - TypeScript SDK
├── cli/ # @fractary/codex-cli - CLI tool
├── mcp/server/ # @fractary/codex-mcp - MCP server
├── plugins/codex/ # Claude Code plugin
│ └── skills/ # Plugin skills (config, sync, memory, etc.)
├── docs/ # Documentation
└── specs/ # Technical specifications
Documentation
- Getting Started Guide - First-time setup walkthrough
- Full Documentation Index - All documentation
- Feature Guides - Organized by function, all interfaces side-by-side
- Configuration Guide - Complete configuration reference
- CLI Command Reference - All commands with options and flags
- SDK API Reference - TypeScript/JavaScript SDK
- MCP Server Reference - AI agent integration
- Plugin Reference - Claude Code plugin
Development
# Install dependencies
npm install
# Build all packages
npm run build
# Run all tests
npm test
# Type check
npm run typecheck
License
Apache-2.0
Установить Codex в Claude Desktop, Claude Code, Cursor
unyly install codexСтавит в Claude Desktop, Claude Code, Cursor и VS Code — сам разбирается с npx, uvx и сборкой из исходников.
Впервые? Поставь CLI: curl -fsSL https://unyly.org/install | sh
Или настроить вручную
Выполни в терминале:
claude mcp add codex -- npx -y @fractary/codexПошаговые гайды: как установить Codex
FAQ
Codex MCP бесплатный?
Да, Codex MCP бесплатный — установка в пару кликов через Unyly без оплаты.
Нужен ли API-ключ для Codex?
Нет, Codex работает без API-ключей и переменных окружения.
Codex — hosted или self-hosted?
Self-hosted: сервер запускается локально на твоей машине командой из раздела установки.
Как установить Codex в Claude Desktop, Claude Code или Cursor?
Открой Codex на unyly.org, выбери вкладку своего клиента (Claude Desktop, Claude Code, Cursor) и нажми Install — конфиг сгенерируется автоматически, без правки JSON.
Изменения
Версии и запрашиваемые доступы со временем.
- Новая версия опубликована
- Новая версия опубликована
- Новая версия опубликована
- Новая версия опубликована
- Новая версия опубликована
- Новая версия опубликована
- Новая версия опубликована
- Новая версия опубликована
Похожие MCP
Opencode Omniroute Plugin
OpenCode plugin for the OmniRoute AI Gateway. Drives dynamic model discovery, /connect auth flow, and multi-instance OmniRoute providers via the official @openc
автор: GitHub ActionsRoblox Studio
Enables AI coding tools to control Roblox Studio for workspace exploration, instance manipulation, and script management. It provides tools for playtesting, sce
автор: paralovAntigravity Drawio
MCP server that enables AI assistants to create, parse, render, and validate Draw.io diagrams programmatically.
автор: S-SUJAN-SFetch
Web content fetching and conversion for efficient LLM usage.
AWS KB Retrieval
Retrieval from AWS Knowledge Base using Bedrock Agent Runtime.
автор: modelcontextprotocolSpring AI MCP Server
Provides auto-configuration for setting up an MCP server in Spring Boot applications.
llm-analysis-assistant
A very streamlined mcp client that supports calling and monitoring stdio/sse/streamableHttp, and can also view request responses through the /logs page. It also
автор: xuzexin-hzMCP-Agent
A simple, composable framework to build agents using Model Context Protocol by [LastMile AI](https://www.lastmileai.dev)
автор: lastmile-aiSpring AI MCP Client
Provides auto-configuration for MCP client functionality in Spring Boot applications.
mcp.natoma.ai
A Hosted MCP Platform to discover, install, manage and deploy MCP servers by [Natoma Labs](https://www.natoma.ai)
Compare Codex with
Не уверен что выбрать?
Найди свой стек за 60 секунд
Автор?
Embed-бейдж для README
Похожее
Все в категории ai
