Mop Flow
БесплатноНе проверенMOP Flow — portable agent memory core, workflow engine, and provider bridge for Claude, Codex, Gemini, and Antigravity.
Описание
MOP Flow — portable agent memory core, workflow engine, and provider bridge for Claude, Codex, Gemini, and Antigravity.
README
MOP Flow
MOP portable AI MemoryCore for Claude, Codex / ChatGPT, Gemini, and Antigravity.
English | Bahasa Melayu
What Is MOP Flow?
MOP Flow is a portable MOP (Memory of Planet) core for AI coding workspaces. It gives every supported AI provider the same project memory, agent rules, workflow gates, artifact folders, autosycn flow, and deployment setup.
The goal is simple: install once, then Claude, Codex / ChatGPT, Gemini, and Antigravity can enter the same project with the same source of truth.
MOP Flow
MOP Flow is the provider-neutral orchestration and skill bridge for MOP. It keeps the MOP brand and rules above upstream Ruflo / Claude Flow runtime compatibility, so Claude, Codex, Gemini, and Antigravity can see the same skill inventory and MCP runtime surface.
npx mop-flow
The active improvement roadmap lives in .MOP/flow/ROADMAP.md.
Install
Run this inside your terminal to launch the interactive Dashboard:
npx mop-flow
The interactive TUI will allow you to Install, Update, Delete, or check the Status of MOP Flow in your project directly without needing to remember complex commands.
After install, open your AI coding chat in that project and run:
/mop-setup
What Gets Installed
| Path | Purpose |
|---|---|
.MOP/ |
MOP state, protocol, scripts, workflow config, and artifact templates. |
AGENTS.md |
Provider-neutral instructions for Codex / ChatGPT coding agents. |
CLAUDE.md |
Claude Code entrypoint and behavior rules. |
GEMINI.md |
Gemini CLI entrypoint. |
.agents/ |
Antigravity-compatible agents and skills. |
.codex/, .gemini/, .claude/ |
Provider-specific config and skill surfaces. |
Core Features
| Feature | What it does |
|---|---|
| Auth Gate | First action is always setup/login. No work starts before the gate passes. |
| Agent Router | 32 specialized agents available via .MOP/STATE.json. Picks one primary agent and adds support agents only when useful. |
| Party Mode | Shows visible agent-to-agent discussion for multi-role decisions. |
| MOP Flow | Provider-neutral skill bridge and MCP runtime wrapper across Claude, Codex, Gemini, and Antigravity. |
| Brain Link Service | Links projects to MOP Agent, registers them locally, and keeps relays alive in the background. |
| MOP Workflow | BMAD-inspired 10-phase flow from idea to release with readiness gates. |
| Artifacts | Keeps plans, specs, reviews, and release notes under .MOP/artifacts/. |
| Autosync | Commits and pushes as the real user, not as the AI tool identity. |
| Auto Deploy | Optional setup path for GitHub, Docker, and Vercel. |
First Session Flow
1. AI reads .MOP/STATE.json
2. If not initialized, AI asks you to run /mop-setup
3. Setup asks for project name, owner, codename, password, mode, language, and GitHub info
4. After login, every task goes through the Agent Router
5. Complex tasks use MOP Workflow and readiness gates before coding
Useful Commands
npx mop-flow
Local project helpers:
node .MOP/scripts/mop-core.mjs status
node .MOP/scripts/mop-core.mjs validate
node .MOP/scripts/mop-flow.mjs status
node .MOP/scripts/mop-flow.mjs link <https://agent/v1/api/link/key>
node .MOP/scripts/mop-flow.mjs service install --start
node .MOP/scripts/mop-flow.mjs service list
node .MOP/scripts/mop-workflow.mjs help --actor <codename> --task "<task>"
node .MOP/scripts/mop-autosycn.mjs run --actor <codename> --reason "<what changed>"
node .MOP/scripts/mop-mcp.mjs start # Native Model Context Protocol (MCP) server
Background Brain Relay
mop-flow link saves the private link token in .MOP/link.json and registers
the project in a per-user service registry on that PC. Run this once per PC to
start all registered project relays automatically when Windows starts or when a
Linux user session starts:
npx mop-flow service install --start
After that, every project you link from the same PC is picked up by the service.
Use npx mop-flow service list to see registered projects. The service registry
does not store link tokens; each project keeps its own gitignored
.MOP/link.json.
npx mop-flow also includes a compact Control Center menu: Install, Update,
Doctor, Status, Link, Delete, Skills, and Exit. The Link menu action is
all-in-one: paste the Brain URL, link the project, then install/start the
background relay service automatically.
Team Workflow
In team mode, work is pushed to a user branch first:
dev/<codename> -> BURHAN-MOP review -> main
BURHAN-MOP acts as the merge guardian. It checks the branch, validates state, and merges only when the workflow is safe.
Release
| Item | Value |
|---|---|
| npm package | mop-flow |
| latest command | npx mop-flow install |
| legacy alias | npx burhan-mop install |
| GitHub release | v1.4.4 |
| Node | >=20 |
Links
- npm: https://www.npmjs.com/package/mop-flow
- GitHub: https://github.com/BURHANDEV-ENTERPRISE/mop-flow
- Bahasa Melayu README: README.bm.md
Установить Mop Flow в Claude Desktop, Claude Code, Cursor
unyly install mop-flowСтавит 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 mop-flow -- npx -y mop-flowFAQ
Mop Flow MCP бесплатный?
Да, Mop Flow MCP бесплатный — установка в пару кликов через Unyly без оплаты.
Нужен ли API-ключ для Mop Flow?
Нет, Mop Flow работает без API-ключей и переменных окружения.
Mop Flow — hosted или self-hosted?
Self-hosted: сервер запускается локально на твоей машине командой из раздела установки.
Как установить Mop Flow в Claude Desktop, Claude Code или Cursor?
Открой Mop Flow на unyly.org, выбери вкладку своего клиента (Claude Desktop, Claude Code, Cursor) и нажми Install — конфиг сгенерируется автоматически, без правки JSON.
Похожие MCP
Fetch
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-hzCompare Mop Flow with
Не уверен что выбрать?
Найди свой стек за 60 секунд
Автор?
Embed-бейдж для README
Похожее
Все в категории ai
