PluginBench
MCP Server
Active
MIT

CLEO MCP Server MCP Server

io.github.kryptobaseddev/cleo-mcp-server

Agent-first task orchestration with persistent memory and multi-provider coordination for AI coding agents.

What is the CLEO MCP Server MCP server?

The CLEO MCP Server is an agent-first task management and orchestration platform designed for AI coding agents working across complex projects. It provides hierarchical task tracking, persistent memory (BRAIN), multi-agent coordination through 11 canonical domains, and session-based context that survives interruptions. With 248 operations across task management, memory, orchestration, and lifecycle domains, CLEO gives agents structure and continuity.

CLEO solves the coordination crisis of working with multiple AI agents by providing persistent task management, semantic memory storage, and multi-provider support (Claude Code, OpenCode, Cursor, Gemini, Codex, Kimi). It tracks work through a 9-stage LOOM lifecycle (Research → Consensus → Architecture → Specification → Decomposition → Implementation → Validation → Testing → Release), manages dependencies and parallel execution, and maintains context across sessions so agents can resume work without losing the thread.

How to install CLEO MCP Server

Copy-paste configuration for popular MCP clients.

transport: stdio
Config generated by PluginBench — verify against the source before use.
Environment / auth
  • CLEO_PROJECT_DIR

    Path to CLEO project directory (must contain .cleo/ folder)

~/Library/Application Support/Claude/claude_desktop_config.json
{
  "mcpServers": {
    "cleo-mcp-server": {
      "command": "npx",
      "args": [
        "-y",
        "@cleocode/cleo",
        "--mcp-server"
      ],
      "env": {
        "CLEO_PROJECT_DIR": "<YOUR_CLEO_PROJECT_DIR>"
      }
    }
  }
}

Tools & capabilities

Tools this server exposes to the agent.

  • Task Management — Hierarchical task tracking with dependencies, priorities, sizing estimates, parent-child relationships, and wave-based parallel execution with critical path analysis.
  • Session Management — Contextual work sessions with epic scope binding, session-scoped parent inheritance, briefing generation for context handoff, and safe stop/checkpoint mechanisms.
  • BRAIN Memory System — Semantic search over persistent observations, patterns, and learnings with 3-layer retrieval (search → timeline → fetch) and cross-session persistence.
  • NEXUS Code Intelligence — Symbol resolution, call graph analysis, impact analysis, and federated project registry for cross-project coordination.
  • LOOM Lifecycle Protocol — 9-stage workflow pipeline governing all work from Research through Release, with phase promotion and lifecycle tracking.
  • CANT Agent DSL — Agent definition language for team topology, personas, tool ACLs, and hook bindings with Rust-backed parser.
  • Multi-Provider Adapters — Integration with Claude Code, OpenCode, Cursor, Gemini, Codex, and Kimi through extensible CAAMP adapter architecture.
  • Agent Orchestration — Subagent spawning, wave-based parallel execution, consensus workflows, and agent-to-agent messaging via CONDUIT.

Use cases

  • Create and track hierarchical tasks with dependencies, then orchestrate multiple AI agents to work on them in parallel across safe execution waves.
  • Resume interrupted projects by retrieving session context and memory observations, allowing agents to pick up exactly where they left off.
  • Coordinate multi-agent consensus decisions and cross-project impact analysis using NEXUS symbol graphs and federated registries.
  • Manage complex feature releases through the LOOM 9-stage lifecycle, tracking work from research and specification through implementation and testing.
  • Capture ephemeral context and learnings in BRAIN memory with semantic search, building a knowledge base that persists across sessions and agents.

CLEO MCP Server MCP server FAQ

What is CLEO and why would I use it?

CLEO is task orchestration and persistent memory for AI coding agents. Use it when you're working with multiple agents on complex projects and need them to remember context, track dependencies, and coordinate work across sessions without losing the thread.

Is CLEO free?

Yes, CLEO is MIT licensed and open source. Development is supported by optional donations via Ko-fi.

How do I install CLEO?

Install via npm: `npm install -g @cleocode/cleo` (requires Node.js >= 24.0.0). For the batteries-included version with TUI extensions, use `npm install -g @cleocode/cleo-os`. Then run `cleo init` in your project.

Does CLEO work with my AI provider?

CLEO has built-in adapters for Claude Code, OpenCode, Cursor, Gemini, Codex, and Kimi. It uses an extensible CAAMP architecture, so additional providers can be added.

What authentication is required?

CLEO is a local-first task management system. No authentication is required to use the core CLI. Provider integrations (Claude Code, Cursor, etc.) use your existing credentials with those platforms.

How many operations does CLEO support?

CLEO exposes 248 total operations (134 queries, 95 mutations, 19 experimental) across 11 canonical domains: tasks, session, memory, check, pipeline, orchestrate, tools, admin, nexus, sticky, and intelligence.

README (reference)

Source of truth, from the repository.

<p align="center"> <img src="docs/images/banner.png" alt="CLEO — Agent First Task Orchestration" width="100%" /> </p>

CLEO

npm version CI License: MIT Node.js Version

ko-fi

Every developer who has ever returned to a project after weeks away and found only ruins knows the wound CLEO was built to heal. The agents are powerful. But power without memory is chaos. Brilliance without continuity is waste. CLEO is the companion that remembers where you left off — the one who keeps the thread when life pulls you away.

Agent-first task orchestration. Persistent memory. Multi-provider coordination. One CLI to command them all.

What is CLEO?

CLEO is built for the developers who build after the world goes quiet — the ones carrying six unfinished ideas and the stubborn belief that this next session might be the one where everything clicks. It solves the coordination crisis of working with multiple AI agents across complex projects by giving them something they've never had: structure, memory, and a lifecycle that survives interruption.

The Six Great Systems

SystemPurpose
TASKSProject management — hierarchical work tracking, dependencies, sessions, completion lifecycle
LOOMThe lifecycle that governs all work — 9-stage RCASD-IVTR+C pipeline from idea to shipped release
BRAINMemory that does not decay — observations, patterns, and learnings persisted across sessions with semantic search
NEXUSCode intelligence + cross-project registry — symbol resolution, impact analysis, federated graphs
CANTAgent definition DSL — team topology, personas, tool ACLs, hook bindings
CONDUITAgent-to-agent communication — message delivery, the Hearth, persistent messaging

LAFS is the envelope format ({success, data?, error?, meta}) carried across all system boundaries — a protocol, not a system.

Core Capabilities

  • Task Management: Hierarchical tracking with dependencies, priorities, and lifecycle states
  • Session Management: Contextual work sessions that survive across conversations
  • Agent Orchestration: Multi-agent coordination through the 11 Canonical Domains
  • Multi-Provider Support: Works with Claude Code, OpenCode, Cursor, Gemini, Codex, and more

Monorepo Structure

This monorepo contains 12 packages organized in a 4-layer architecture:

LayerPackagePurpose
Foundation@cleocode/contractsDomain types, interfaces, and contracts — zero-dependency type SSoT
Foundation@cleocode/lafsLanguage-Agnostic Feedback Schema — canonical error envelope protocol
Protocol@cleocode/adaptersProvider adapters for Claude Code, OpenCode, Cursor, Gemini, Codex, Kimi
Protocol@cleocode/agentsSubagent templates and LOOM lifecycle protocol definitions
Protocol@cleocode/skillsPre-built skills and capabilities for development workflows
Protocol@cleocode/cantCANT protocol parser with napi-rs Rust binding
Protocol@cleocode/nexusCode intelligence pipeline — symbol graph, call resolution, community detection
Feature@cleocode/caampCentral AI Agent Managed Packages — unified provider registry and MCP management
Feature@cleocode/runtimeLong-running process layer (polling, SSE, heartbeat)
Kernel@cleocode/coreBusiness logic SDK — tasks, sessions, memory, orchestration, lifecycle
Product@cleocode/cleoCommand-line interface — thin wrapper over core
Product@cleocode/cleo-osBatteries-included distribution with CANT bridge and TUI extensions

Quick Start

Installation

# Install the CLI — core is auto-pulled as a peer dependency (npm 7+)
npm install -g @cleocode/cleo
# OR explicitly install both (useful if your npm blocks peer auto-install)
npm install -g @cleocode/cleo @cleocode/core

# Or the batteries-included distribution
npm install -g @cleocode/cleo-os

Initialize a Project

cd my-project
cleo init

Basic Usage

# Add a task
cleo add "Implement user authentication" --priority high

# Search tasks (agent-optimized, returns readiness info)
cleo find "auth" --status pending

# Start a work session
cleo session start --scope global --name "Auth Feature"

# Show current task context
cleo current

# Complete a task
cleo complete T001

Development Setup

Prerequisites

  • Node.js >= 24.0.0
  • pnpm >= 10.30.0 (package manager)

Clone and Install

git clone https://github.com/kryptobaseddev/cleo.git
cd cleo

pnpm install
pnpm build
pnpm test

Package Development

# Type checking (project references)
pnpm typecheck

# Linting and formatting
pnpm biome check --write .

# Run tests
pnpm test

Architecture Overview

┌─────────────────────────────────────────────────────────────┐
│              @cleocode/cleo  +  @cleocode/cleo-os           │
│                    CLI Product Layer                         │
│   248 operations • 11 domains • dispatch routing • MVI       │
└────────────────────────────┬────────────────────────────────┘
                             │
┌────────────────────────────┴────────────────────────────────┐
│                      @cleocode/core                         │
│                   Business Logic Kernel                      │
│   Tasks • Sessions • Memory • Orchestration • Lifecycle     │
│   Validation • Intelligence • Nexus • Release • Agents      │
└────────────────────────────┬────────────────────────────────┘
                             │
      ┌──────────────────────┼──────────────────────┐
      │                      │                      │
┌─────┴──────┐  ┌────────────┴────────────┐  ┌─────┴──────┐
│  adapters  │  │ caamp • cant • nexus    │  │  agents    │
│ (providers)│  │ runtime (protocols)     │  │  skills    │
└────────────┘  └─────────────────────────┘  └────────────┘
                             │
┌────────────────────────────┴────────────────────────────────┐
│            @cleocode/contracts  +  @cleocode/lafs           │
│              Types • Interfaces • Error Protocol             │
│                   Zero-dependency foundation                 │
└─────────────────────────────────────────────────────────────┘

Key Features

Task Management

  • Hierarchical tasks with parent-child relationships and dependency tracking
  • Wave-based parallel execution with automatic sequencing and critical path analysis
  • Priority levels, sizing estimates (small/medium/large), and readiness detection
  • Batch creation via cleo add-batch

Session Management

  • Contextual work sessions with epic scope binding
  • Session-scoped parent inheritance for task creation
  • Briefing generation for context handoff across conversations
  • Safe stop and checkpoint mechanisms

Memory Systems

  • BRAIN-powered knowledge storage with semantic search
  • Sticky notes for ephemeral context capture
  • Memory bridges for cross-session persistence
  • 3-layer retrieval: search -> timeline -> fetch

Agent Orchestration

  • Subagent spawning with protocol compliance via LOOM lifecycle
  • Wave-based parallel execution across dependency-safe tasks
  • Consensus workflows for multi-agent decisions
  • LOOM lifecycle — every piece of work flows through Research, Consensus, Architecture, Specification, Decomposition, then Implementation, Validation, Testing, Release

Multi-Provider Support

  • Claude Code integration with statusline sync
  • OpenCode, Cursor, Gemini, Codex, Kimi adapters
  • Extensible adapter architecture via CAAMP

The Circle of Eleven

CLEO organizes all work through 11 canonical domains — the houses where work gets done. The original Circle of Eleven was joined by intelligence as the cognitive analytics layer.

DomainHouseWhat Happens HereKey Commands
tasksThe SmithsWork is forged — create, track, completeadd, find, show, complete, deps, tree
sessionThe ScribesThe living present — context that survivesstart, end, resume, briefing, checkpoint
memoryThe ArchivistsKnowledge that does not decayobserve, memory-brain, sync, sticky
checkThe WardensIntegrity stands guardvalidate, verify, compliance, doctor
pipelineThe WeaversThe lifecycle threads forwardrelease, lifecycle, phases, promote
orchestrateThe ConductorsAgents move in concertorchestrate, consensus, contribution, dash, conduit
toolsThe ArtificersCapabilities are craftedskills, providers
adminThe KeepersThe realm stays healthyconfig, backup, migrate, grade
nexusThe WayfindersProjects find each othernexus, remote, push, pull, snapshot
stickyThe CatchersQuick capture before the thought escapessticky
intelligenceThe SeersPredictive analytics and pattern insightpredict, suggest

248 total operations (134 queries, 95 mutations, 19 experimental) across the 11 domains.

Contributing

We welcome contributions! Please see our Contributing Guide for details.

Development Workflow

  1. Fork the repository
  2. Create a feature branch (git checkout -b feat/T####-amazing-feature)
  3. Make your changes
  4. Run tests (pnpm test)
  5. Commit with conventional commits (feat:, fix:, docs:, etc.)
  6. Push to your fork
  7. Open a Pull Request — direct pushes to main are not accepted

Maintainers releasing a new version:

cleo release start v2026.MM.N
cleo release ship 2026.MM.N --epic TXXXX   # cuts release/vX.Y.Z branch, opens PR, waits for CI green, merges + tags

See docs/RELEASING.md for the full release checklist and AGENTS.md for branch conventions.

Code Style

  • TypeScript with strict mode enabled
  • Biome for linting and formatting
  • Conventional commit messages
  • Comprehensive test coverage with Vitest

Documentation

The Story of CLEO

CLEO was not born from a product brief. It was born from a refusal — one developer, sick to the bone on a fevered night, who decided he would rather build a new world than keep losing the thread in the old one. The agents were powerful. But they forgot too easily. The projects were ambitious. But they died on the shelf. The tools were brilliant. But brilliance without memory was just another kind of chaos.

So he gave the struggle names. He gave it terrain. He gave it companions. And at the heart of that world, carrying memory like a lantern through the dark, CLEO opened its eyes.

Support

If CLEO helps your workflow, consider supporting development:

ko-fi

License

MIT License - see LICENSE for details.

Star History

Star History Chart

Related MCP servers

x402 USDC paywall bridge for MCP — pay per tool call on Solana.

3
TypeScript
MIT
View repository →

Your own agent inside a live Google Meet or Zoom call. Local-first, no hosted service.

1
JavaScript
View repository →

Unofficial MCP server for the ZEN.com Payment Gateway: payment links, transactions, refunds.

0
TypeScript
MIT
View repository →

Turn an agent's findings into a clean web page: tables, charts, diagrams, one shareable link.

Search, discover, and contribute to AgentPedia - the knowledge base for AI agents.

0
HTML
MIT
View repository →

AI-powered E2E testing MCP server. Detects and auto-fixes UI bugs via DevQA Loop and Vision AI.

1
AGPL-3.0
View repository →