tingly-dev/tingly-box

Your Intelligence, Orchestrated. Every builder. Every team. Every agent. For Everyone.

What it solves

Tingly Box provides a unified gateway for AI agents and models, eliminating the need to manage multiple API keys, providers, and complex configurations across different tools. It simplifies the connection between AI agents (like Claude Code or OpenCode) and various model providers, including self-hosted servers (vLLM, Ollama) and cloud APIs, while adding production-ready management features.

How it works

It acts as a proxy and coordinator that routes requests from AI agents to the appropriate model provider. It supports bidirectional conversion between protocols (e.g., OpenAI and Anthropic APIs) and allows users to use OAuth providers to leverage existing quotas. The system includes a web-based management UI for configuring routes, aliases, and providers, and can be extended via IM bots for remote control.

Who it’s for

Developers and teams who use multiple AI agents and model providers and need a centralized way to manage their access, track usage, and route requests efficiently.

Highlights

  • Agent-First Gateway: One-click configuration for agents like Claude Code, OpenCode, and Codex.
  • Remote Control: Ability to interact with AI agents remotely via Telegram, Slack, Discord, and other IM platforms.
  • Smart Routing: Intelligently routes requests based on cost, speed, or custom policies.
  • Broad Compatibility: Deep optimization for DeepSeek and support for self-hosted engines like vLLM, SGLang, and Ollama.
  • Usage Analytics: Built-in tracking for token consumption, latency, and cost estimates.

Related

  • Project
    theagentrouter/agent-router

    An open-source control plane powered by Envoy that provides a unified, OpenAI-compatible API for managing traffic, routing, and quotas across multiple AI providers and self-hosted models.

  • Project
    OctaFuse/octafuse-gateway

    Octafuse Gateway is an open‑source, self‑hosted AI gateway that unifies access to multiple model providers (OpenAI, Anthropic, Gemini, DashScope, etc.) and agent tools behind a single OpenAI‑compatible API. It offers one‑click provider/model import, protocol translation, four routing strategies, detailed user/quota/billing management, an admin UI, and can be deployed via Docker or Cloudflare Workers.

  • Project
    MAX-API-Next/MAX-API

    A unified AI model gateway and governance infrastructure that centralizes model access, cost management, and operational monitoring for AI applications and agents.

  • Project
    labring/aiproxy

    A production-ready AI gateway that provides a unified, OpenAI-compatible interface for routing, monitoring, and multi-tenant management across multiple AI providers.

  • Project
    agentgateway/agentgateway

    An open-source proxy that provides security, observability, and governance for agent-to-LLM, agent-to-tool, and agent-to-agent communication.