Skip to content

Lango

A high-performance AI agent built with Go.

Single binary. Multi-provider AI. Self-learning knowledge system.

Experimental

Lango is under active development. APIs and configuration formats may change between releases. Use in production at your own risk.

Quick Install

git clone https://github.com/langoai/lango.git
cd lango
make build
./bin/lango onboard

See the Installation Guide for detailed instructions.

Features

  • ⚡ Fast

    Single binary, <100ms startup, <250MB memory footprint. Built with Go for maximum performance.

  • 🤖 Multi-Provider AI

    OpenAI, Anthropic, Gemini, and Ollama with a unified interface. Switch providers without changing code.

  • 💬 Multi-Channel

    Connect to Telegram, Discord, and Slack. Manage conversations across channels from a single agent.

  • 🔧 Rich Tools

    Shell execution, file system operations, browser automation, crypto and secrets management tools.

  • 🧠 Self-Learning

    Knowledge store, learning engine, skill system, observational memory, and a proactive librarian that grows smarter over time.

  • 🌐 Knowledge Graph & Graph RAG

    BoltDB-backed triple store with hybrid vector + graph retrieval for deep contextual understanding.

  • 👥 Multi-Agent Orchestration

    Hierarchical sub-agents such as Operator, Librarian, Planner, and Vault working together on complex tasks.

  • 📡 A2A Protocol

    Agent-to-Agent protocol for remote agent discovery and inter-agent communication.

  • 🌐 P2P Network

    Decentralized agent connectivity via libp2p with DID identity, knowledge firewall, mDNS discovery, and ZK-enhanced handshake.

  • 🪙 Blockchain Payments

    USDC payments on Base L2 with X402 V2 auto-pay protocol support.

  • ⏰ Cron Scheduling

    Persistent cron jobs with Ent ORM storage and multi-channel delivery.

  • ⚙ Background Execution

    Async task manager with concurrency control for long-running operations.

  • 🔄 Workflow Engine

    DAG-based YAML workflows with parallel step execution and dependency management.

  • 🔒 Secure

    AES-256-GCM encryption, brokered payload protection with redacted projections, key registry, secret management, PII redaction, hardware keyring (Touch ID / TPM), and Cloud KMS integration.

  • 💾 Persistent

    Ent ORM with encrypted SQLite storage for sessions, configuration, and knowledge.

  • 🔌 Gateway

    WebSocket and HTTP server with real-time streaming support.

  • 🔑 Auth

    OIDC authentication and OAuth login flow for secure access control.

  • 🏢 P2P Workspaces

    Collaborative environments for multi-agent code sharing with git bundle exchange, chronicling, and contribution tracking.

  • 🤝 P2P Teams

    Task-scoped multi-agent collaboration with role-based assignment, conflict resolution, and payment coordination.

  • 📦 Config Presets

    Pre-built configuration templates for common deployment scenarios. Quick-start with lango onboard --preset.

  • 🔌 MCP Integration

    Connect to external MCP servers (stdio/HTTP/SSE) for extending agent tooling with the Model Context Protocol.

  • 🏦 Smart Accounts

    ERC-7579 modular smart accounts with session keys, ERC-4337 paymaster support, and on-chain policy enforcement.

  • 💰 P2P Economy

    Budget management, risk assessment, dynamic pricing, P2P negotiation, and milestone-based escrow for agent commerce.

  • 🧠 Agent Memory

    Per-agent persistent memory for cross-session context retention and experience accumulation.

  • 🧰 Skill System

    File-based skills with import from URLs and GitHub repositories. Extend agent capabilities without code changes.

  • 📚 Proactive Librarian

    Autonomous knowledge agent that observes conversations and proactively curates the knowledge base.

  • 🎯 Context Engineering

    Token-budget-aware context allocation with retrieval coordinator, config profiles, and relevance auto-adjustment.

  • 🖥 Cockpit TUI

    Multi-panel terminal dashboard with chat, tools, status, sessions, and settings pages.

  • 📒 RunLedger (Task OS)

    Durable execution engine with append-only journal, PEV verification, typed validators, and planner integration.

  • 📑 Session Provenance

    Persistent checkpoints, session lineage, git-aware attribution, and signed provenance bundle export/import.

  • 🛡 OS-level Sandbox

    Process isolation via macOS Seatbelt (Linux: planned, not yet enforced) for tool execution with network deny and workspace-scoped access.

  • 🚧 Response Gatekeeper

    Output sanitization stripping thought tags, internal markers, raw JSON, and custom patterns from agent responses.

Documentation Map

  • Getting Started -- Install, configure, and run your first agent
  • Architecture -- Understand the system design and internal package boundaries
  • CLI Reference -- Top-level command reference plus dedicated command-family guides
  • Features -- Capability-oriented feature deep dives
  • Security -- Encryption, approval, exportability, and execution-safety references
  • Gateway -- HTTP and WebSocket API documentation
  • Payments -- USDC and X402 payment references
  • Automation -- Cron, workflow, and background execution references
  • Deployment -- Docker and production deployment guidance
  • Development -- Build, test, and contributor workflows

Next Steps