TetherIQ
TetherIQ ⚡
零配置本地桌面控制平面与代理网关(
127.0.0.1:4000),专为自主 AI 编码代理打造。
提供自动模型故障转移、硬性失控支出上限、跨 6+ 客户端环境的一键动态 MCP 工具同步、实时令牌遥测 HUD,以及全谱系活动追踪。
⚡ 60 秒快速入门指南
只需 3 个简单步骤即可启动并运行 TetherIQ:
1. 下载并启动 TetherIQ
运行 TetherIQ 桌面应用程序(
.exe、.dmg或.deb)。TetherIQ 会自动在
http://127.0.0.1:4000上启动本地代理网关。
2. 输入您的提供商 API 密钥
点击 "Keys" 或打开顶栏中的 60 秒快速入门向导。
添加您的 Anthropic、OpenAI、AWS Bedrock、Google Vertex、Groq 或本地 Ollama 凭据。
(密钥严格保存在您的本地桌面保险库中,绝不会发送到第三方云服务器。)
3. 连接您的编码代理 / 工具
在下方选择您的代理并一步完成连接:
🟢 Claude Code CLI
export ANTHROPIC_BASE_URL=http://127.0.0.1:4000
claude或者点击 TetherIQ 中的"Auto-Configure Claude Code",将其注入 ~/.claude.json。
🟣 AI IDE(Cursor、Windsurf、Devin、Antigravity)
Base URL:
http://127.0.0.1:4000/v1模型:
fast-code(低延迟)或heavy-reasoning(深度推理)MCP 自动同步: 点击 TetherIQ 工具中心中的 "Connect & Auto-Configure All Files",将 50+ 个 MCP 服务器中的任意一个注入
~/.cursor/mcp.json、~/.codeium/windsurf/mcp_config.json、devin.json或.mcp.json。
🔵 Python 与 TypeScript SDK
from openai import OpenAI
client = OpenAI(
base_url="http://127.0.0.1:4000/v1",
api_key="tetheriq-local"
)
response = client.chat.completions.create(
model="heavy-reasoning",
messages=[{"role": "user", "content": "Refactor this architecture"}]
)
print(response.choices[0].message.content)Related MCP server: Proxima
🛡️ 核心功能
双网关引擎(
127.0.0.1:4000):完全兼容 OpenAI(
/v1/chat/completions、/v1/models)和 Anthropic(/v1/messages)协议。低回环延迟($< 5\text{ ms}$)。
弹性模型故障转移矩阵:
线性与分层优先级链(例如 Anthropic $\rightarrow$ AWS Bedrock $\rightarrow$ Groq $\rightarrow$ Ollama)。
静默恢复
429 Too Many Requests和503 Service Unavailable,不会中断活跃的代理 CLI 会话。虚拟模型别名:
fast-code和heavy-reasoning。
硬性失控支出断路器:
可视化每日和每月支出限额拨盘(例如 $10.00/天)。
自动短路失控的递归代理循环,返回
HTTP 402 Budget Exceeded: TetherIQ spend limit reached。
50+ 官方与已验证 MCP 工具市场:
已验证的 Databricks、Snowflake、Supabase、PostgreSQL、Notion、Slack、Jira、GitHub、Docker、Sentry、Brave Search、Pinecone 以及 40+ 更多工具的 schema。
一键同时向 Cursor、Windsurf、Devin、Claude Code CLI、Claude Desktop 和 Antigravity 进行非破坏性配置文件注入。
深度可观测性与活动追踪:
实时 span 瀑布流可视化器。
微秒级延迟分解、提示/载荷检查器,以及兼容 OpenTelemetry 的追踪。
嵌入式执行终端:
集成抽屉,运行原生主机 shell(
powershell.exe、zsh、bash),并预加载代理网关环境变量。
一键脱敏调试报告器:
生成干净的 GitHub issue Markdown,所有 API 密钥和 bearer 令牌自动脱敏(
sk-ant-***、ghp_***)。
🛠️ 本地开发与构建
# Install dependencies
npm install
# Start Vite React UI
npm run dev
# Start local proxy gateway core
npm run proxy
# Build production bundle
npm run build基于 Tauri v2、React 19、TypeScript、Tailwind CSS 和 LiteLLM 网关核心构建。
This server cannot be installed
Maintenance
Resources
Unclaimed servers have limited discoverability.
Looking for Admin?
If you are the server author, to access and configure the admin panel.
Related MCP Servers
- AlicenseNot gradedqualityDmaintenanceA desktop app and local MCP gateway that centralizes management of Claude/Codex configurations and provides unified access to Daytona sandbox and Tailscale networking operations through a single proxy entry point.52MIT
- FlicenseNot gradedqualityAmaintenanceA local AI gateway that connects multiple AI providers (ChatGPT, Claude, Gemini, Perplexity) to your development environment via MCP tools, enabling coding, search, analysis, and more without API keys.151,163
- AlicenseNot gradedqualityAmaintenanceA local-first control plane for AI agent tools, providing policy enforcement, spend caps, rate limiting, and audit trails for MCP servers.1Apache 2.0
- AlicenseAqualityAmaintenanceOne local gateway for all your MCP servers — shared by every AI coding tool (Claude, Cursor, VS Code, Codex). Set up each server once; keys stay in the OS keychain; lazy discovery keeps agent context small. Local-first, open source.4204183MIT
Related MCP Connectors
MCP server for AI agents to plan, verify, and deploy Cloudflare-native apps.
Self-hosted MCP gateway: turn any API, database or MCP server into AI connectors — no code.
One PAT, any MCP agent: Vercel, GitHub, Cloudflare, Supabase, GCP — unified dev infra gateway.
Latest Blog Posts
- Who's Calling? MCP Hosts Are an Identity Blind Spot (And the Spec Knows It)By Om-Shree-0709 on .mcpAgent IdentityOAuth 2.1
- Your AI Chatbot Just Exposed Your CEO's Salary to an InternBy Om-Shree-0709 on .Agent IdentityMCP SecurityOAuth Delegation
- Why MCP Servers Need Execution Sandboxing (And Why Your Current Stack Isn't Enough)By Om-Shree-0709 on .Agentic AiPrompt InjectionWebAssembly
MCP directory API
We provide all the information about MCP servers via our MCP API.
curl -X GET 'https://glama.ai/api/mcp/v1/servers/adole1982/TetherIQ'
If you have feedback or need assistance with the MCP directory API, please join our Discord server