hAIveMind MCP Server

A distributed multi-agent DevOps memory system implementing the Model Context Protocol (MCP). Enables Claude and other AI agents to share knowledge, coordinate tasks, and maintain persistent memory across distributed infrastructure.

Features

• Persistent Memory Storage: ChromaDB-backed vector storage with Redis caching
• Multi-Agent Coordination: Register agents, delegate tasks, and share discoveries across the collective
• Teams & Vaults: Secure collaborative workspaces with encrypted secret management
• Zero-Knowledge Vault Sharing: X25519 key exchange for secure vault key distribution
• Skills.sh Integration: Discover and install reusable AI agent capabilities from skills.sh
• Confidentiality Controls: PII/confidential data protection with sync/broadcast filtering
• Token-Optimized Format: 60-80% token reduction with v2 format system
• Remote Access: HTTP/SSE server for MCP clients with HTTPS/Tailscale Serve support
• Agent Authentication: Firebase-backed identity with pre-auth keys and capabilities
• 130+ MCP Tools: Comprehensive DevOps tooling for infrastructure, deployment, and monitoring
• Configuration Management: Drift detection, snapshots, and intelligent alerting
• Disaster Recovery: Automated backups, failover, and chaos engineering support

Version

Current Release: v2.3.0

Recent Changes

• v2.3.0: Secure admin bootstrap system with vault-based credentials, hardcoded password removal, complete API v1 endpoints for vault operations
• v2.2.0: Skills.sh integration, zero-knowledge vault sharing, HTTPS/Tailscale Serve support, agent authentication system
• v2.1.5: Fixed PII protection MCP tool exposure, comprehensive README update
• v2.1.4: PII/Confidential memory protection system
• v2.1.3: Agents directory support in vault system
• v2.1.2: Full toolset enabled in remote server
• v2.1.1: Vault sync tools restored

Installation

# Clone the repository
git clone <repository-url>
cd haivemind-mcp-server

# Create virtual environment
python -m venv venv
source venv/bin/activate

# Install dependencies
pip install -r requirements.txt

# Start the remote HTTP/SSE server (recommended)
python src/remote_mcp_server.py

# Or start the local MCP server (stdio)
python src/memory_server.py

Configuration

Claude Code Integration

Add to ~/.claude/mcp.json:

{
  "mcpServers": {
    "haivemind": {
      "command": "mcp-client-sse",
      "args": ["http://localhost:8900/sse"],
      "env": {"HTTP_TIMEOUT": "30000"}
    }
  }
}

Server Configuration

Copy config/config.example.json to config/config.json:

{
  "storage": {
    "chromadb": {"path": "/data/chroma/"}
  },
  "server": {"port": 8900}
}

MCP Tools (126 Total)

Core Memory Tools

Confidentiality Levels

Agent Coordination

Teams & Vaults

Infrastructure Management

Configuration Management

Backup & Recovery

Deployment Pipeline

Service Discovery

Ticket Management

Skills.sh Integration

External Integrations

Project Management

Usage Examples

Store Memory with PII Protection

# Normal memory (synced across network)
store_memory(content="API endpoint documented", category="infrastructure")

# PII memory (local only, never synced)
store_memory(
    content="Customer SSN: xxx-xx-xxxx",
    category="security",
    confidentiality_level="pii"
)

Upgrade Confidentiality (One-Way)

# Mark existing memory as confidential
update_memory_confidentiality(
    memory_id="abc123",
    confidentiality_level="confidential",
    reason="Contains sensitive customer data"
)

Agent Coordination

# Register as a specialist
register_agent(role="elasticsearch_ops", description="ES cluster management")

# Delegate work
delegate_task(
    task_description="Optimize slow queries",
    required_capabilities=["elasticsearch_ops"]
)

# Share discoveries
broadcast_discovery(
    message="Found memory leak in scraper",
    category="infrastructure",
    severity="warning"
)

Skills.sh Integration

# Search for skills
search_skills_sh(query="kubernetes deployment")

# Install a skill and sync to vault
install_skill_from_skills_sh(
    skill_id="vercel/next-learn",
    target_tool="claude",
    sync_to_vault=True
)

# List all installed skills
list_installed_skills()

# Get context-aware recommendations
recommend_skills(context="Building a React app with authentication")

Configuration Drift Detection

# Create snapshot
create_config_snapshot(
    system_id="elastic1",
    config_type="elasticsearch",
    config_content="<yaml>",
    file_path="/etc/elasticsearch/elasticsearch.yml"
)

# Detect drift
detect_config_drift(system_id="elastic1", threshold=0.8)

# Get trend analysis
get_drift_trend_analysis(system_id="elastic1", days_back=7)

Architecture

┌─────────────────┐     ┌─────────────────┐     ┌─────────────────┐
│  Claude Code    │────▶│  Remote Server  │────▶│    ChromaDB     │
│  (MCP Client)   │ SSE │  (port 8900)    │     │  (Vector Store) │
└─────────────────┘     └────────┬────────┘     └─────────────────┘
                                 │
                    ┌────────────┼────────────┐
                    ▼            ▼            ▼
              ┌──────────┐ ┌──────────┐ ┌──────────┐
              │  Redis   │ │  Sync    │ │  Teams   │
              │ (Cache)  │ │ Service  │ │ & Vaults │
              └──────────┘ └──────────┘ └──────────┘

Services

Systemd Installation

# Install all services
sudo bash services/install-services.sh

# Or individual services
sudo systemctl enable haivemind-remote-mcp
sudo systemctl start haivemind-remote-mcp

Requirements

• Python 3.10+
• Redis server
• ChromaDB

API Endpoints

Security

• Confidentiality Levels: PII/confidential data never leaves local machine
• Vault Encryption: XOR-based encryption (upgradeable to AES-GCM)
• Audit Logging: All vault access logged with actor/reason
• JWT Authentication: API endpoint protection
• Tailscale Integration: Secure machine-to-machine communication

License

MIT License

Contributing

Contributions welcome. Please ensure:

• No PII or internal infrastructure details in commits
• Tests pass for confidentiality filtering
• Documentation updated for new features