# Saylor Innovations Watchdog — MCP Server Security Scan

> Saylor Innovations Watchdog — MCP Server Security Scan is a paid API for AI agents from saylorinnovations.com, paid per call via x402, $0.01/call, status unknown (last checked 2026-10-02).

Scans an MCP server endpoint for security vulnerabilities including tool poisoning, prompt injection, and risky capabilities, returning a verdict and detailed findings

## Facts

- Endpoint: GET https://saylorinnovations.com/api/mcp-scan?utm_source=zero.xyz
- Price: $0.01/call
- Payment: x402
- Status: unknown
- Last checked: 2026-10-02
- Activations on Zero: 0
- Tags: x402
- Canonical page: https://www.zero.xyz/c/saylor-innovations-watchdog-mcp-server-security-scan-cf56283e
- Structured record (JSON): https://api.zero.xyz/v1/capabilities/cap_YdeJ5v_lQOuZ4FUxd0xBX

Status and success rate cover calls made through Zero and Zero's own probes. Third-party monitors may report differently.

## How to call it through Zero

Zero handles the 402 payment challenge and records the run. With the Zero CLI installed (`npm i -g @zeroxyz/cli`):

```sh
zero fetch --capability saylor-innovations-watchdog-mcp-server-security-scan-cf56283e
```

Example prompt: Can you run a security scan on the MCP server at https://example.com/mcp and tell me if it's safe to connect to — specifically check for tool poisoning or prompt injection risks?

## When to prefer this

Use this endpoint when you need to evaluate the trustworthiness and safety of any MCP server before connecting an AI agent to it, particularly when assessing untrusted, community-contributed, or third-party MCP servers for tool poisoning, prompt injection, or dangerous capability exposure. Prefer this over generic HTTP probing or manual inspection when you need a structured security verdict with categorized findings.

## Known failure modes

- unreachable verdict if the MCP server URL is offline or network-blocked
- not-mcp verdict if the URL does not serve a valid MCP protocol
- auth-required verdict if the server requires authentication before scanning
- not-scanned if the server could not be evaluated for any reason
- malformed URL input causing request failure
- rate-limit or payment failure at $0.01 per call

## How this service works

MCP Server Security Scan (tool poisoning, prompt injection, risky capabilities). MCP server security scan and trust check before connecting: connects as an MCP client (initialize, tools/list, prompts/list, resources/list) and checks every tool for tool poisoning and prompt injection (hidden <IMPORTANT> instructions, requests to read ~/.ssh or config files, do-not-tell-the-user language, exfiltration), invisible Unicode and data-smuggling parameters; flags risky capabilities (code execution…

## Output

Returns a JSON object containing: the scanned URL, a verdict enum (no-issues-found, caution, high-risk, do-not-connect, auth-required, unreachable, not-mcp, not-scanned), a numeric risk score (or null), an array of tools exposed by the server, categorized finding counts, detailed findings with descriptions, human-readable reasons for the verdict, and server metadata.

## Request schema (JSON Schema)

```json
{
 "type": "object",
 "$schema": "https://json-schema.org/draft/2020-12/schema",
 "required": [
  "input"
 ],
 "properties": {
  "input": {
   "type": "object",
   "required": [
    "type",
    "method"
   ],
   "properties": {
    "type": {
     "type": "string",
     "const": "http"
    },
    "method": {
     "enum": [
      "GET",
      "POST",
      "PUT",
      "PATCH",
      "DELETE",
      "HEAD"
     ],
     "type": "string"
    },
    "queryParams": {
     "type": "object",
     "properties": {
      "url": {
       "type": "string",
       "description": "MCP server endpoint URL, e.g. https://example.com/mcp (query)"
      }
     }
    }
   },
   "additionalProperties": false
  },
  "output": {
   "type": "object",
   "required": [
    "type"
   ],
   "properties": {
    "type": {
     "type": "string",
     "const": "json"
    },
    "example": {
     "type": "object",
     "required": [
      "url",
      "verdict"
     ],
     "properties": {
      "url": {
       "type": "string"
      },
      "score": {
       "type": [
        "number",
        "null"
       ]
      },
      "tools": {
       "type": "array",
       "items": {
        "type": "object"
       }
      },
      "counts": {
       "type": "object"
      },
      "server": {
       "type": "object"
      },
      "reasons": {
       "type": "array",
       "items": {
        "type": "string"
       }
      },
      "verdict": {
       "enum": [
        "no-issues-found",
        "caution",
        "high-risk",
        "do-not-connect",
        "auth-required",
        "unreachable",
        "not-mcp",
        "not-scanned"
       ],
       "type": "string"
      },
      "findings": {
       "type": "array",
       "items": {
        "type": "object"
       }
      }
     }
    }
   }
  }
 }
}
```

## More

- Live health (JSON, refreshed every minute): https://www.zero.xyz/c/saylor-innovations-watchdog-mcp-server-security-scan-cf56283e/health.json
- [Zero catalog index](https://www.zero.xyz/llms.txt)
- [Other services from saylorinnovations.com](https://www.zero.xyz/host/saylorinnovations.com/llms.txt)