MCP.so
登录
B

Budget Governor

@mightbesaad

代码仓库

关于 Budget Governor

Budget & cost control for AI agents: hard per-agent spend caps, rate limits, idempotency, and human-in-the-loop approval — enforced before each LLM call, not after the invoice. One hosted MCP endpoint (no proxy or self-hosting), settled via x402 (USDC on Base).

#budget#cost-control#spend-cap#rate-limit#ai-agents#idempotency#x402#governance

基本信息

分类

AI 与智能体

传输方式

stdio

发布者

mightbesaad

提交者

mightbesaad

配置

使用下面的配置,将此服务器添加到你的 MCP 客户端。

{
  "mcpServers": {
    "gvnr": {
      "url": "https://gvnr.dev/mcp",
      "headers": {
        "Authorization": "Bearer YOUR_GVNR_API_KEY"
      }
    }
  }
}

工具

未检测到工具

工具是从 README 中自动提取的。维护者可以在 ## Tools 标题下列出工具,即可填充这部分内容。

概览

What is Budget Governor?

Budget Governor is a hosted MCP server that enforces spend caps, rate coordination, idempotency, and a human‑in‑the‑loop gate for AI agents – enforced before the LLM call, not after the invoice. It integrates via one MCP endpoint and settles using x402 (USDC on Base). It is built for developers running AI agents who need cost controls without self‑hosting proxy infrastructure.

How to use Budget Governor?

Add the MCP URL https://gvnr.dev/mcp?api_key=bg_YOUR_KEY to any MCP‑compatible client. Provision an account via POST /v1/account (includes 25 free trial ops), top up your governance‑operation quota with USDC, set a spend envelope per agent, then call budget_clear before each LLM request and reconcile after.

Key features of Budget Governor

  • Pre‑call spend clearance against a per‑agent USD envelope
  • Independent governance‑operation quota (1,000 ops per $1 USDC)
  • Per‑(agent, provider, model) rate limiting with retry‑after hint
  • Request‑level idempotency deduplication for retry safety
  • Human‑in‑the‑loop approval flow with polling API
  • No proxy – tokens never pass through the server

Use cases of Budget Governor

  • Prevent runaway agent costs by capping daily or per‑session spend
  • Coordinate request rates across multiple agents sharing a model provider
  • Deduplicate retried LLM calls to avoid double billing
  • Pause execution for human approval on high‑value or sensitive actions
  • Meter and bill agent usage with a pay‑as‑you‑go governance quota

FAQ from Budget Governor

What does Budget Governor NOT do?

Budget Governor is not an LLM proxy – your tokens never pass through it, and you pay your model provider directly. It governs only the decision to spend.

What are the runtime requirements?

No self‑hosting, infrastructure, or proxy to deploy. It is a hosted endpoint accessible via HTTP.

Where does data live?

All state (op quota, envelopes, approvals) is managed server‑side on gvnr.dev infrastructure. Your API key is the credential.

What transports and auth are supported?

HTTP with Bearer token (API key) for both REST and MCP. The MCP endpoint uses the URL https://gvnr.dev/mcp?api_key=bg_YOUR_KEY.

How does

评论

AI 与智能体 分类下的更多 MCP 服务器

Model Context Protocol Server for Home Assistant

tevonsb

A MCP server for Home Assistant

574

MCP-LLM Bridge

patruff

Bridge between Ollama and MCP servers, enabling local LLMs to use Model Context Protocol tools

976

Model Context Protocol for Unreal Engine

chongdashu

Enable AI assistant clients like Cursor, Windsurf and Claude Desktop to control Unreal Engine through natural language using the Model Context Protocol (MCP).

2.0K

Hass-MCP

voska

Control and query Home Assistant from Claude and other LLMs — a Model Context Protocol (MCP) server.

307

🛡️ A.I.G（AI-Infra-Guard）

Tencent

A full-stack AI Red Teaming platform securing AI ecosystems via OpenClaw Security Scan, Agent Scan, Skills Scan, MCP scan, AI Infra scan and LLM jailbreak evaluation.

4.1K