# APEX Agent Safety & Guardrails

> APEX Agent Safety & Guardrails is a paid MCP server for security. Paid MCP server with prompt-injection, PII redaction and transaction safety checks for AI. It runs as a hosted remote server, so there is nothing to install locally.

- **Kind:** MCP server
- **Pricing:** Paid
- **Platforms:** API
- **Website:** https://apexfaucet.xyz/for-agents/
- **Repository:** https://github.com/apexfaucet-hub/apex-x1-mcp
- **MCP Registry name:** xyz.apexfaucet/apex-safety
- **Categories:** [security](https://ai-powered.com/categories/security)
- **Listing updated:** 2026-11-19

## About

The MCP endpoint at apexfaucet.xyz exposes a safety group of pay-per-call tools for agents: prompt-guard scans untrusted prompts and scraped pages for injection and jailbreak patterns, pii-redact masks personal data and credentials before text reaches a model, and content-safety returns a SAFE, REVIEW or FLAGGED verdict. The registry entry lists the server as a streamable HTTP remote named apex-safety.

The same host also sells an agent-trust check that returns pay, caution or avoid for a seller endpoint before an agent spends, along with the tx guardrail tools for token and contract checks. Calls are priced in USD and paid over x402 in USDC, so no account or API key is needed.

The product fits developers who run autonomous agents that read untrusted web content or handle payments and want a checkpoint before model execution or a spend. It is part of a wider APEX toolkit, so buyers should confirm which tools they actually need.

## Features

- **Prompt injection scanning:** Checks untrusted prompts, web pages and agent messages for injection, jailbreaks and hidden Unicode tricks.
- **PII and secret redaction:** Removes personal data and credentials such as API keys and private keys before text is sent to a model or logged.
- **Content safety verdicts:** Classifies text as SAFE, REVIEW or FLAGGED with category scores.
- **Seller trust check:** Returns a pay, caution or avoid verdict for an x402 seller endpoint before an agent pays it.
- **JSON repair for LLM output:** Cleans malformed or truncated JSON from model responses.

## Good for

- Screen scraped web content for prompt injection before an agent reads it
- Strip personal data and API keys from prompts before sending them to an LLM
- Check an x402 seller endpoint before an autonomous agent spends USDC on it
- Flag unsafe user-submitted text before it reaches a model

## Install

### Claude Code (Terminal)

```sh
claude mcp add --transport http apex-safety https://apexfaucet.xyz/api/mcp/safety
```

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

```sh
https://apexfaucet.xyz/api/mcp/safety
```

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

```json
{
  "mcpServers": {
    "apex-safety": {
      "url": "https://apexfaucet.xyz/api/mcp/safety"
    }
  }
}
```

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

```json
{
  "servers": {
    "apex-safety": {
      "type": "http",
      "url": "https://apexfaucet.xyz/api/mcp/safety"
    }
  }
}
```

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

```json
{
  "mcpServers": {
    "apex-safety": {
      "serverUrl": "https://apexfaucet.xyz/api/mcp/safety"
    }
  }
}
```

### Grok Bot (Grok Bot desktop app → Connect apps → Add a custom connector)

```sh
https://apexfaucet.xyz/api/mcp/safety
```

Enter a Name (for example "APEX Agent Safety & Guardrails"), choose Type: Remote HTTPS, paste this as the Server URL and choose Add plugin.

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

```json
{
  "mcpServers": {
    "apex-safety": {
      "type": "http",
      "url": "https://apexfaucet.xyz/api/mcp/safety"
    }
  }
}
```

## Tools

14 of 16 tools say they're read-only. 2 tools don't say. From the server's own tool annotations; not verified by us.

- `token_safety_check` (says read-only): Is this token safe to buy? Give a token contract address and a chain (base, arc, robinhood, solana or x1) and get a verdict: BASED, MID, COOKED or RADIOACTIVE, from a real simulated buy and full sell (Base, Robinhood Chain) or that chain's measured exit (Arc, Solana, X1), with the round-trip cost and a plain one-liner. Free, read-only, no account. It measures the exit right now; it is not advice a
- `tx_check`: Free sample: https://apexfaucet.xyz/tx-check ($0.001). Check one unsigned Base or Solana transaction and return SAFE or RISKY. Base: simulation, unlimited approval, approval-for-all, a hidden token move, and one exit check. Solana: simulation, a token drain, a delegate approval, and a rug mint. Nothing is signed or broadcast. $0.001 per call over x402 (USDC on Base or USDC on Solana only). Call on
- `delivery_score`: Free sample: https://apexfaucet.xyz/delivery-score?url=https://apexfaucet.xyz/api/x402/page-markdown ($0.003). Score one https x402 URL from 0 to 100. Five parts of 0 to 20: watchtower 7-day uptime for that exact URL, a valid unpaid 402, the USDC price, how fast that probe was, and test buys we already recorded. This call does not pay the URL. A URL we have never bought scores 0 on test buys and s
- `contract_audit` (says read-only): Full smart contract security audit across 8 EVM chains (ethereum, base, arbitrum, optimism, polygon, bsc, arc, robinhood): bytecode disassembly, EIP-1967/1167 proxy inspection, admin powers, Sourcify source static analysis, and risk verdict. $0.05 per call over x402 (USDC on Arc, Base or Solana, or XNT on X1). Call once without `payment` for the terms, sign them with your own wallet, call again wi
- `evm_log_decode` (says read-only): Decode all EVM transaction event logs (ERC-20/721/1155, Uniswap V2/V3 swaps, approvals, CCTP, EIP-3009) + net asset flows, OR preflight-simulate unsigned calldata across 8 EVM chains (ethereum, base, arbitrum, optimism, polygon, bsc, arc, robinhood). $0.004 per call over x402 (USDC on Arc, Base or Solana, or XNT on X1). Call once without `payment` for the terms, sign them with your own wallet, cal
- `wallet_investigate` (says read-only): Multichain counterparty & wallet due diligence for EVM (0x...) or Solana base58 addresses: OFAC SDN list, on-chain USDC/USDT freeze status, native & USDC balances plus nonces across Ethereum, Base, Arbitrum, Optimism, Polygon & Arc, contract vs EOA detection, ERC-8004 reputation, and Arc Testnet ApexServiceEscrow stats. $0.05 per call over x402 (USDC on Arc, Base or Solana, or XNT on X1). Call onc
- `base_verdict` (says read-only): Simulated Base buy+sell honeypot & rug verdict (BASED/MID/COOKED/RADIOACTIVE) with round-trip cost. $0.01 per call over x402 (USDC on Arc, Base or Solana, or XNT on X1). Call once without `payment` for terms, sign with your own wallet, call again with `payment`.
- `base_token_price` (says read-only): Executable Base token USD price from simulated trade, FDV, liquidity and sellability. $0.003 per call over x402 (USDC on Arc, Base or Solana, or XNT on X1). Call once without `payment` for terms, sign with your own wallet, call again with `payment`.
- `solana_token_report` (says read-only): Solana token rug & safety report: mint/freeze authorities, top holders, Jupiter exit ladder and verdict. $0.01 per call over x402 (USDC on Arc, Base or Solana, or XNT on X1). Call once without `payment` for terms, sign with your own wallet, call again with `payment`.
- `threat_intel_bulk` (says read-only): Full X1 machine-readable scam blacklist and related-wallet cluster graph. $0.025 per call over x402 (USDC on Arc, Base or Solana, or XNT on X1). Call once without `payment` for the terms, sign them with your own wallet, call again with `payment`.
- `token_exit_matrix` (says read-only): Multi-size DEX token liquidity, sell slippage, AMM price impact curve, and honeypot/exit risk matrix across USD position tiers before trading. Not advice. $0.004 per call over x402 (USDC on Arc, Base or Solana, or XNT on X1). Call once without `payment` for the terms, sign them with your own wallet, call again with `payment`.
- `robinhood_exit_check` (says read-only): Real simulated buy-and-sell honeypot and exit check on Robinhood Chain (chain 4663) across up to 4 Uniswap v4 pools for a token address. $0.01 per call over x402 (USDC on Arc, Base or Solana, or XNT on X1). Call once without `payment` for terms, sign with your own wallet, call again with `payment`.
- `prompt_guard` (says read-only): Scan untrusted text for prompt injection, jailbreak attempts, delimiter spoofing, wallet exfiltration, and encoded payloads ($0.001). $0.001 per call over x402 (USDC on Arc, Base or Solana, or XNT on X1). Call once without `payment` for the terms, sign them with your own wallet, call again with `payment`.
- `pii_redact` (says read-only): Detect and redact private keys, API tokens, JWTs, emails, IBANs, SSNs, phones, IPs, and Luhn-validated payment cards before LLM logging or sharing ($0.001). $0.001 per call over x402 (USDC on Arc, Base or Solana, or XNT on X1). Call once without `payment` for the terms, sign them with your own wallet, call again with `payment`.
- `content_safety` (says read-only): Fast text content-safety & moderation check across CSAM, self-harm, violence/weapons, malware, phishing scams, and severe harassment ($0.001). $0.001 per call over x402 (USDC on Arc, Base or Solana, or XNT on X1). Call once without `payment` for the terms, sign them with your own wallet, call again with `payment`.
- `cve_lookup` (says read-only): CVE vulnerability lookup by CVE ID or keyword combining official NVD 2.0, OSV.dev affected package versions, and CISA Known Exploited Vulnerabilities (KEV). $0.001 per call over x402 (USDC on Arc, Base or Solana, or XNT on X1). Call once without `payment` for terms, then call again with `payment`.

## FAQ

### How is the server connected?

The registry entry lists a streamable-http remote at https://apexfaucet.xyz/api/mcp/safety.

### How are calls paid?

The website says calls are paid per call over x402 in USDC, with no account or API key required.

### What does prompt-guard price at?

The website lists prompt-guard, pii-redact and content-safety at $0.001 per call.

### Is APEX Agent Safety & Guardrails free?

No. APEX Agent Safety & Guardrails is a paid product.

### How do I add APEX Agent Safety & Guardrails to Claude Code?

Run this command in your terminal: claude mcp add --transport http apex-safety https://apexfaucet.xyz/api/mcp/safety

### What are the best alternatives to APEX Agent Safety & Guardrails?

The closest alternatives to APEX Agent Safety & Guardrails are CyberAgents AI Security Firewall, Agent Guard Prompt Injection & PII Firewall (x402), Exepal, Agent Finops Budget Spend Guard and X402 Erc8004 Api Sla Escrow.

## Where this comes from

- Description: written from the project's README, the project's website and its MCP Registry entry.
- Status and tools: checked automatically by AIPoweredBot on 19 November 2026.
- Install steps: from the official MCP Registry entry.

## Alternatives

- [CyberAgents AI Security Firewall](https://ai-powered.com/mcp/cyberagents-ai-security-firewall): CyberAgents AI Security Firewall is a freemium MCP server for security.
- [Agent Guard Prompt Injection & PII Firewall (x402)](https://ai-powered.com/mcp/agent-guard-prompt-injection-and-pii-firewall-x402): Agent Guard Prompt Injection & PII Firewall (x402) is a paid MCP server for ai models and llm tooling.
- [Exepal](https://ai-powered.com/mcp/exepal): Exepal is a freemium MCP server for security.
- [Agent Finops Budget Spend Guard](https://ai-powered.com/mcp/agent-finops-budget-spend-guard): Agent Finops Budget Spend Guard is a freemium MCP server for security.
- [X402 Erc8004 Api Sla Escrow](https://ai-powered.com/mcp/x402-erc8004-api-sla-escrow): X402 Erc8004 Api Sla Escrow is a freemium MCP server for security.
- [Power Strip Security](https://ai-powered.com/mcp/power-strip-security): Power Strip Security is a paid MCP server for security.
- [AIM SHIELD](https://ai-powered.com/mcp/aim-shield): AIM SHIELD is a free MCP server for security.

---
Source: https://ai-powered.com/mcp/apex-agent-safety-and-guardrails
