Connect Pyxmark to your AI agent

To use Pyxmark in your AI agent, add https://mcp.pyxmark.com/mcp as a remote MCP server (Streamable HTTP, no sign in) with the steps for your client below.

Server address (Streamable HTTP, no sign in or key):

https://mcp.pyxmark.com/mcp

Pyxmark is a hosted MCP server, so there’s nothing to install and no key to paste. Free use is 25 lookups a day per IP address, and every tool call counts as one. Once it’s connected, ask “List the Pyxmark tools” to check it works.

ChatGPT

On the web, go to ChatGPT Plugins at chatgpt.com/plugins, select the plus button, then Add custom MCP server. Name it Pyxmark, enter the URL as the Server URL, choose No authentication, select I understand and want to continue, then Create as a plugin. Install it and pick it with @ in a chat.

In the ChatGPT desktop app: Settings, then MCP servers, then Add server. Choose Streamable HTTP, paste the URL, save, then select Restart.

Step by step: Pyxmark in ChatGPT.

Claude (claude.ai and Claude Desktop)

Customize → Connectors → + Add → Add custom connector. Name it Pyxmark, paste the URL and click Continue. Review the authentication settings and click Continue, choose No sign in, then finish by clicking Add.

Step by step: Pyxmark in Claude.

VS Code

With GitHub Copilot in VS Code, run MCP: Add Server from the Command Palette, choose HTTP (HTTP or Server-Sent Events) and paste the URL. Enter pyxmark as the server ID, then select Global, and then Copilot Global when prompted. Or from a terminal:

code --add-mcp '{"name":"pyxmark","type":"http","url":"https://mcp.pyxmark.com/mcp"}'

Step by step: Pyxmark in Cursor and VS Code.

Cursor

Add this to ~/.cursor/mcp.json (all projects) or .cursor/mcp.json (one project):

{ "mcpServers": { "pyxmark": { "url": "https://mcp.pyxmark.com/mcp" } } }

Step by step: Pyxmark in Cursor and VS Code.

Claude Code

claude mcp add --transport http pyxmark https://mcp.pyxmark.com/mcp

Step by step: Pyxmark in Claude.

Codex

codex mcp add pyxmark --url https://mcp.pyxmark.com/mcp

The ChatGPT desktop app and the Codex IDE extension share this setting.

Step by step: Pyxmark in ChatGPT.

Gemini CLI

gemini mcp add --transport http -s user pyxmark https://mcp.pyxmark.com/mcp

With -s user it’s added for all your projects; without it, only the current one. Check it with gemini mcp list, or type /mcp inside Gemini CLI.

From the Gemini CLI MCP docs.

Perplexity

Open Plugins in the sidebar (or Customize → Connectors if you don’t see Plugins), then Add custom → Connector → Add MCP connector, and select Remote. Name it Pyxmark, enter the URL as the MCP Server URL, set Authentication to None and Transport to Streamable HTTP. Check the acknowledgement box and click Add, then click the connector card to turn it on.

In an organization, members can add custom connectors only if an admin allows it.

From Perplexity’s guide to adding custom remote connectors.

Microsoft Copilot Studio

On your agent’s Tools page, select Add a tool, then New tool, then Model Context Protocol. Enter Pyxmark as the Server name, a short Server description such as “Sanctions screening and vendor checks”, and the URL as the Server URL. Choose None for authentication and select Create. Then on Add tool, select Create a new connection, and then Add to agent.

From Microsoft’s guide to connecting an agent to an existing MCP server.

Grok

Go to grok.com/connectors, click New Connector, then select Custom, and enter the URL as the MCP server URL. On Grok Business and Enterprise a team admin adds it first.

Step by step: Pyxmark in Grok.

Devin Desktop (legacy Cascade agent)

Windsurf’s MCP docs now point to Devin Desktop. These steps are for the legacy Cascade agent only. Devin Local, the default agent for new tabs, keeps its MCP servers in separate config files.

Enterprise users must manually turn MCP on via settings. Then in the Cascade panel, click the … (Actions) menu in the top right, then Open MCP config file in the MCPs section, and add:

{ "mcpServers": { "pyxmark": { "serverUrl": "https://mcp.pyxmark.com/mcp" } } }

The file is ~/.config/devin/mcp_config.json, or %APPDATA%\devin\mcp_config.json on Windows.

From the Devin Desktop MCP docs.

JetBrains AI Assistant

Go to Settings | Tools | AI Assistant | Model Context Protocol (MCP) and click Add. In the New MCP Server dialog, choose Connect to the server over HTTP, paste this JSON, click OK, then Apply:

{ "mcpServers": { "pyxmark": { "url": "https://mcp.pyxmark.com/mcp" } } }

From the AI Assistant MCP docs.

Amazon Q Developer

In the IDE, open the Q Developer Chat panel, choose the tools icon, then the plus (+) symbol. Pick global or local scope, enter pyxmark as the Name, select http as the transport, enter the URL, leave Headers empty and choose Save.

From the Q Developer MCP docs.

Grok Build

grok mcp add --transport http pyxmark https://mcp.pyxmark.com/mcp

Step by step: Pyxmark in Grok.

Grok Bot

Ask Grok Bot in chat:

Add this MCP server: https://mcp.pyxmark.com/mcp

Step by step: Pyxmark in Grok.

Cline

Click the MCP Servers icon in the top toolbar of the Cline panel and open the Remote Servers tab. Enter pyxmark as the Server Name and the URL as the Server URL, choose Streamable HTTP as the Transport Type and click Add Server. Or add this to cline_mcp_settings.json:

{ "mcpServers": { "pyxmark": { "type": "streamableHttp", "url": "https://mcp.pyxmark.com/mcp", "disabled": false } } }

Keep the type: without it, Cline falls back to the older SSE transport.

From the Cline MCP docs.

Continue

Continue uses MCP servers in agent mode. Create a .continue/mcpServers folder at the top of your workspace and add a YAML file with the lines below. The name, version and schema fields at the top are required in a file of its own.

name: Pyxmark
version: 0.0.1
schema: v1
mcpServers:
  - name: pyxmark
    type: streamable-http
    url: https://mcp.pyxmark.com/mcp

From the Continue MCP docs.

Goose

Run goose configure, select Add Extension, then Remote Extension (Streamable HTTP), and follow the prompts with pyxmark as the name and the URL. To use it for one session only:

goose session --with-streamable-http-extension "https://mcp.pyxmark.com/mcp"

From the Goose extensions docs.

Other clients

Most clients read the standard block below. Claude Code and Grok Build also pick it up from a project’s .mcp.json.

{ "mcpServers": { "pyxmark": { "type": "http", "url": "https://mcp.pyxmark.com/mcp" } } }

Clients that only run local stdio servers can use:

npx -y mcp-remote https://mcp.pyxmark.com/mcp

See mcp-remote. AI agents setting Pyxmark up for someone can follow llms-install.md.

Limits

Results are informational, not a compliance determination or legal advice, and names are matched fuzzily, so check each potential match against the linked official entry. How matching, verdicts and list refreshes work: sources, method and limits.