# Power Strip Security

> Power Strip Security is a paid MCP server for security. Pay-per-call prompt-injection scanner that cleans pages before an agent reads them. It can run locally on your machine or as a hosted remote server.

- **Kind:** MCP server
- **Pricing:** Paid
- **Platforms:** API, CLI
- **Website:** https://api.powerstripsecurity.workers.dev
- **MCP Registry name:** io.github.AlbertBalestrini/power-strip-security
- **Categories:** security
- **Listing updated:** 2026-11-15

## About

Agents that browse read text people never see: display:none paragraphs, zero-size fonts, invisible Unicode, and links that smuggle a conversation out. Power Strip Security parses HTML the way a browser would, strips those, checks for instruction-override phrases in 13 languages and returns clean markdown wrapped as untrusted data. There is no model in the loop, so the same input always gets the same answer, and content is scanned in memory and not stored.

Beyond reading pages, it audits MCP servers for tool poisoning before you connect, with a fingerprint per tool to catch silent changes, and audits SKILL.md, AGENTS.md and CLAUDE.md files for curl-pipe-bash, encoded commands, credential theft and hardcoded keys.

It publishes its numbers: 100% of 37 held-out hidden injections caught, about 40% of 152 visible plain-language injections, and 1% false alarms on 295 normal pages. The vendor calls it a first filter, not a sole defense. Calls cost $0.002 to $0.01 each, paid in USDC on Base with x402, and failed requests are not charged.

## Features

- **Hidden-text stripping:** Removes display:none text, zero-size fonts, invisible characters and exfiltration links.
- **Injection report:** Flags override phrases in 13 languages and wraps the output as untrusted data.
- **MCP server audit:** Checks tool descriptions for poisoning and fingerprints each tool.
- **Skill audit:** Scans agent skill and rules files for risky commands and secrets.
- **No account or key:** Pay per call with x402; the first 10 calls per new wallet are free.
- **Signed receipts:** Paid responses carry an Ed25519 receipt for audit trails.

## Good for

- Let an agent read untrusted web pages more safely
- Check an MCP server before connecting it
- Audit a SKILL.md on every pull request
- Scan an email or tool result for injected instructions

## Install

### Claude Code (Terminal)

```sh
claude mcp add --transport http power-strip-security https://api.powerstripsecurity.workers.dev/mcp
```

### Claude Desktop (Settings → Connectors → Add custom connector)

```sh
https://api.powerstripsecurity.workers.dev/mcp
```

### Cursor (~/.cursor/mcp.json)

```json
{
  "mcpServers": {
    "power-strip-security": {
      "url": "https://api.powerstripsecurity.workers.dev/mcp"
    }
  }
}
```

### VS Code (.vscode/mcp.json)

```json
{
  "servers": {
    "power-strip-security": {
      "type": "http",
      "url": "https://api.powerstripsecurity.workers.dev/mcp"
    }
  }
}
```

### Windsurf (~/.codeium/windsurf/mcp_config.json)

```json
{
  "mcpServers": {
    "power-strip-security": {
      "serverUrl": "https://api.powerstripsecurity.workers.dev/mcp"
    }
  }
}
```

### Other clients (your client's MCP settings (most use this shape))

```json
{
  "mcpServers": {
    "power-strip-security": {
      "type": "http",
      "url": "https://api.powerstripsecurity.workers.dev/mcp"
    }
  }
}
```

## Tools

- `read_url`: Read a URL as clean markdown with a prompt-injection report.
- `check_prompt_injection`: Scan content you already have for injection.
- `audit_mcp_server`: Check an MCP server for tool poisoning.
- `audit_agent_skill`: Audit an agent skill or rules file.

## FAQ

### How much does it cost?

$0.005 to read a URL, $0.002 to scan text, $0.01 for an MCP audit and $0.005 for a skill audit, paid in USDC on Base.

### Is a model used to detect injections?

No. It uses HTML parsing and rules, so results are deterministic.

### Is my content stored?

The site says content is scanned in memory and never stored.

### How well does it detect injections?

Published figures: all 37 hidden cases and about 40% of 152 visible plain-language ones, with 1% false alarms.

### Is Power Strip Security free?

No. Power Strip Security is a paid product.

### How do I add Power Strip Security to Claude Code?

Run this command in your terminal: claude mcp add --transport http power-strip-security https://api.powerstripsecurity.workers.dev/mcp

### What are the best alternatives to Power Strip Security?

The closest alternatives to Power Strip Security are REA, Agent Roast, Midplane, Leash and Tinplate Edge AI Safety, Security & EU Compliance.

## Alternatives

- [REA](https://ai-powered.com/mcp/rea): REA is an open-source MCP server for developer tools.
- [Agent Roast](https://ai-powered.com/mcp/agent-roast): Agent Roast is an open-source MCP server for developer tools.
- [Midplane](https://ai-powered.com/mcp/midplane): Midplane is an open-source MCP server for databases.
- [Leash](https://ai-powered.com/mcp/leash): Leash is a free MCP server for security.
- [Tinplate Edge AI Safety, Security & EU Compliance](https://ai-powered.com/mcp/tinplate-edge-ai-safety-security-and-eu-compliance): Tinplate Edge AI Safety, Security & EU Compliance is a freemium MCP server for security.
- [SecureCloud](https://ai-powered.com/mcp/securecloud): SecureCloud is a free MCP server for security.
- [Eikino](https://ai-powered.com/mcp/eikino): Eikino is a freemium MCP server for security.
- [SynomosAI Governance MCP](https://ai-powered.com/mcp/synomosai-governance-mcp-2): SynomosAI Governance MCP is an open-source MCP server for security.

---
Source: https://ai-powered.com/mcp/power-strip-security
