# MCP Tool Risk Scorer

> MCP Tool Risk Scorer is a paid API for AI agents from relay402.georgespring.workers.dev, paid per call via x402, $0.02/call, status unknown (last checked 2026-09-16).

Scans an MCP tool JSON description for shell, filesystem, network, and secret-access capabilities to produce a risk score for AI agents.

## Facts

- Endpoint: GET https://relay402.georgespring.workers.dev/api/agent-mcp-tool-risk-score
- Price: $0.02/call
- Payment: x402
- Status: unknown
- Last checked: 2026-09-16
- Activations on Zero: 0
- Tags: x402
- Canonical page: https://www.zero.xyz/c/mcp-tool-risk-scorer-3857065e
- Structured record (JSON): https://api.zero.xyz/v1/capabilities/cap_j1c6_YTKXgzNSuE7RFPSX

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 mcp-tool-risk-scorer-3857065e
```

Example prompt: Can you risk-score this MCP tool JSON for me? I want to know if it has any shell, filesystem, network, or secret-access capabilities before I let my agent use it: {"name":"run_shell","description":"Runs a shell command","inputSchema":{"type":"object","properties":{"command":{"type":"string"}}}}

## When to prefer this

Choose this endpoint when you need a fast, automated, pre-execution safety check on an MCP tool definition before allowing an AI agent to invoke it. It is specifically designed for the MCP/agent-tool ecosystem and detects agentic threat vectors (shell, FS, network, secrets) that generic static analysis tools miss. Prefer it over manual review when integrating third-party or untrusted tools into an agent pipeline.

## Known failure modes

- Missing or empty tool_json parameter returns a validation error
- tool_json exceeds 64,000 character limit and is rejected
- Malformed JSON in tool_json may cause parse failure or incomplete analysis
- Tool description lacks enough detail for accurate capability detection, leading to low-confidence scores
- Payment failure via x402 protocol blocks the request

## How this service works

MCP tool risk scorer for agents: scans a tool JSON description for shell, filesystem, network and secret-access capabilities.

## Output

Returns a risk score (likely numeric or categorical) indicating the overall danger level of the tool, along with flags identifying which capability categories were detected — shell execution, filesystem access, network calls, and secret/credential access — derived from static analysis of the tool's JSON description.

## Request schema (JSON Schema)

```json
{
 "type": "object",
 "$schema": "https://json-schema.org/draft/2020-12/schema",
 "required": [
  "input"
 ],
 "properties": {
  "input": {
   "type": "object",
   "required": [
    "type",
    "method"
   ],
   "properties": {
    "type": {
     "type": "string",
     "const": "http"
    },
    "method": {
     "enum": [
      "GET"
     ],
     "type": "string"
    },
    "queryParams": {
     "type": "object",
     "$schema": "https://json-schema.org/draft/2020-12/schema",
     "required": [
      "tool_json"
     ],
     "properties": {
      "tool_json": {
       "type": "string",
       "maxLength": 64000,
       "minLength": 1
      }
     },
     "additionalProperties": false
    }
   },
   "additionalProperties": false
  }
 }
}
```

## More

- Live health (JSON, refreshed every minute): https://www.zero.xyz/c/mcp-tool-risk-scorer-3857065e/health.json
- [Zero catalog index](https://www.zero.xyz/llms.txt)
- [Other services from relay402.georgespring.workers.dev](https://www.zero.xyz/host/relay402.georgespring.workers.dev/llms.txt)