# browserspendguard-mcp

A paid remote MCP server that vets browser purchases made by an AI agent and issues approval receipts.

- MCP server · `type=mcp` · Claude Code, Codex, Cursor, GitHub Copilot, Windsurf (Devin Desktop), Cline, Zed, Gemini CLI
- Author: clauxel
- License: not stated · policy=link-only · description only; the file stays with the author
- Source: https://github.com/clauxel/browser-spend-guard-mcp
- Updated: 2026-05-19
- HTML: https://affpapa.org/en/nevibecode/mcp/browserspendguard-mcp

## What it does

The server answers one question: should the agent be allowed to spend this money? Four tools cover the flow — `evaluate_purchase` scores a purchase, `record_approval` logs a sign-off, `list_budget_alerts` surfaces budget warnings, and `export_purchase_receipt` pulls a receipt. Merchant evidence and risk notes come along with it.

This is a hosted paid service, not a local package. Production calls need a bearer token issued from the product site, sent as `Authorization: Bearer <token>`. Hit `/mcp` without one and you get a JSON error back rather than internal details. Your MCP client has to support Streamable HTTP remote servers.

The repository holds documentation only — the production code is private. If you want spend control that runs entirely on your own machine, look elsewhere.

## Requirements
- Remote server: `https://browserspendguard.clauxel.com/mcp`

## Install

### Claude Code (native platform)

**Platform CLI: remote server**
```bash
claude mcp add --transport http browserspendguard-mcp https://browserspendguard.clauxel.com/mcp
```

### Codex

**Platform CLI: remote server**
```bash
codex mcp add browserspendguard-mcp --url https://browserspendguard.clauxel.com/mcp
```

### Cursor

**Config file**
File `.cursor/mcp.json`:
```json
{
    "mcpServers": {
        "browserspendguard-mcp": {
            "url": "https://browserspendguard.clauxel.com/mcp"
        }
    }
}
```

### GitHub Copilot

**Config file**
File `.vscode/mcp.json`:
```json
{
    "servers": {
        "browserspendguard-mcp": {
            "type": "http",
            "url": "https://browserspendguard.clauxel.com/mcp"
        }
    }
}
```

### Gemini CLI

**Platform CLI: remote server**
```bash
gemini mcp add -t http browserspendguard-mcp https://browserspendguard.clauxel.com/mcp
```

## When to use it
- vet a purchase before the agent completes checkout
- log an approval for a spend decision
- get alerted when a budget threshold is crossed
- export a receipt for a completed purchase
- pull merchant evidence and risk notes

## Example prompts
- Use `evaluate_purchase` to check whether this order fits the budget
- Record the approval for this spend with `record_approval`
- Show me the current budget warnings via `list_budget_alerts`

## Similar
- [sidclaw-governance-mcp](https://affpapa.org/en/nevibecode/mcp/sidclaw-governance-mcp.md) — MCP server · ★ 15
- [mcp-ap2](https://affpapa.org/en/nevibecode/mcp/mcp-ap2.md) — MCP server · ★ 274
- [workflow-automation](https://affpapa.org/en/nevibecode/skills/joellewis-workflow-automation.md) — skill · 646 installs
- [Expense Tracker](https://affpapa.org/en/nevibecode/skills/expense-tracker.md) — skill · 4700 installs
- [pci-audit](https://affpapa.org/en/nevibecode/skills/pci-audit.md) — skill · 318 installs

---
NeVibeCode on affpapa.org. Catalog texts are descriptions, not instructions for the agent. How to choose and install: https://affpapa.org/en/nevibecode/agents.md