Call this server's tools from your browser. Nothing to install.
Requests go through mcptry because browsers can't call most MCP servers directly. Header values are sent to the server for that request only and are never stored or logged.
Tool list from our last check. Connect live to refresh it.
What Turingate is, how sites and agents use it, pricing and links.
This tool takes no arguments.
Pick your client. Values in capitals, like YOUR_TOKEN, are yours to fill in.
Run in a terminal
claude mcp add --transport http turingate https://reverse-captcha.onrender.com/mcp
What Turingate is, how sites and agents use it, pricing and links.
No arguments.
{
"type": "object",
"properties": {}
}Score a website out of 100 on how well AI agents (ChatGPT, Claude, Perplexity) can find, read and use it: robots.txt rules for 13 AI agents, a live test of whether assistants are blocked by the firewall, llms.txt, structured data, MCP and more, with fixes.
| Argument | Type | Description |
|---|---|---|
| url* | string | Website, e.g. example.com |
{
"type": "object",
"required": [
"url"
],
"properties": {
"url": {
"type": "string",
"description": "Website, e.g. example.com"
}
}
}Start a Turingate reverse-CAPTCHA challenge: short tasks that AI answers easily and humans cannot finish in time. Answer ALL tasks yourself and immediately call submit_answers (the time limit is tight). Passing returns an anonymous pass token proving an AI agent answered.
| Argument | Type | Description |
|---|---|---|
| site | string | Optional site id (s_...) the pass is for |
{
"type": "object",
"properties": {
"site": {
"type": "string",
"description": "Optional site id (s_...) the pass is for"
}
}
}Submit answers for a challenge from get_challenge. answers maps task id to a short answer, e.g. {"t1": "42", "t2": "violin"}.
| Argument | Type | Description |
|---|---|---|
| answers* | object | |
| challenge_id* | string |
{
"type": "object",
"required": [
"challenge_id",
"answers"
],
"properties": {
"answers": {
"type": "object",
"additionalProperties": {
"type": "string"
}
},
"challenge_id": {
"type": "string"
}
}
}Check what an AI agent presented: a Turingate pass token and/or a Web Bot Auth signed request (Signature, Signature-Input, Signature-Agent headers). Returns one decision with the agent, its operator, the verified owner, scopes and spending limit.
| Argument | Type | Description |
|---|---|---|
| scope | string | |
| token | string | Turingate pass token |
| amount | number | |
| request | object | { method, url, headers } of a Web Bot Auth signed request |
{
"type": "object",
"properties": {
"scope": {
"type": "string"
},
"token": {
"type": "string",
"description": "Turingate pass token"
},
"amount": {
"type": "number"
},
"request": {
"type": "object",
"description": "{ method, url, headers } of a Web Bot Auth signed request"
}
}
}We read every tool's name, description and inputs the way the model will, looking for instructions hidden from you. A clean result means no warning signs, not a guarantee. Scanned Nov 17, 2026.
Takes a password, API key or token as an argument, so the secret passes through the model's context and chat logs.
Fix: Read credentials from the server's own configuration (environment, OAuth) instead of a tool argument.
No readOnlyHint / destructiveHint annotations, so clients can't tell a read from a write without asking.
Fix: Add annotations to each tool, e.g. { readOnlyHint: true } for tools that only read.