MCP Server

PingAura runs a remote MCP server. It lets AI assistants query your account data directly. This includes domains, visibility scores, site audits, and articles. You never have to leave your editor or chat interface.

Supported clients include Claude Code, Claude Desktop, Cursor, Codex, VS Code, Windsurf, Perplexity, ChatGPT, Gemini CLI, and Amazon Q.

Connect in 30 seconds

Pick the path that matches where you'll use PingAura.

OAuth — for Claude.ai (recommended)

If you're using Claude on the web or desktop, this is the fastest setup. No keys to copy or paste.

1. In Claude, go to Settings → Connectors → Add custom connector.
2. Paste https://www.pingaura.ai/api/mcp.
3. Claude opens a consent page on pingaura.ai. Sign in if you aren't already.
4. Pick which team account Claude should connect to (if you belong to more than one).
5. Choose what Claude can do: Read your data, or Read + Modify your data (recommended so Claude can act on what it finds — create prompts, run audits, publish articles).
6. Click Approve. Claude is connected, with all 29 tools ready.

You can disconnect any time from Claude's connector settings — that immediately invalidates Claude's access.

Full walk-through: OAuth 2.1 Server.

API key — for Claude Desktop, Cursor, Codex, VS Code, etc.

Anything that takes a header in a config file uses this path. Same access as OAuth, just configured once per machine.

1. In your PingAura dashboard, go to Settings → API Keys.
2. Click Create API Key, name it (e.g. "Cursor on my laptop"), and copy the key. You'll only see it once — store it in a password manager.
3. Drop it into your client's MCP config (see Client setup below) as either header:
   • Authorization: Key pa_xxx_yyy
   • Authorization: Bearer pa_xxx_yyy

Each API key is tied to a single team account, with full read + write access. Only account owners can create or revoke keys; team members can see existing ones.

Quick start

The fastest way to connect is with Claude Code:

claude mcp add pingaura --transport http https://www.pingaura.ai/api/mcp \
  --header "Authorization: Key YOUR_API_KEY"

Then ask Claude something like: "List my domains and their visibility scores"

Quick setup

Paste your API key to install PingAura directly into your editor.

PingAura API Key

Install in CursorInstall in VS CodeCopy Claude Code command

Enter your API key to enable install buttons.

About MCP clients

An MCP client is any application that speaks the Model Context Protocol — desktop apps, IDE extensions, CLI tools, and agent frameworks all qualify. There are 50+ MCP clients available today, and PingAura works with all of them because we implement the standard streamable HTTP transport. You can use any client that supports streamable HTTP transport; just point it at https://www.pingaura.ai/api/mcp and supply your API key in the Authorization header.

Secret headers / API key in config

For clients that let you specify custom headers in a config file, the API key always goes in the Authorization header using one of these two forms:

Authorization: Key pa_xxx_yyy
Authorization: Bearer pa_xxx_yyy

Your API key is sensitive. Store it in an environment variable or a password manager — do not commit it into version-controlled config files. Many clients support env-var interpolation (e.g. $PINGAURA_API_KEY) in their config; use that where available.

Client setup

OAuth or API key? Most modern MCP clients (Claude.ai, Claude Desktop, Claude Code, Cursor, recent versions of others) support OAuth — you point them at our URL and they handle the consent flow in your browser, no key needed. The configs below show both paths where the client supports OAuth: try OAuth first, fall back to API key only if your client requires headers in config or runs in an environment without a browser.

Claude Code (CLI)

With OAuth (no API key needed):

claude mcp add pingaura --transport http https://www.pingaura.ai/api/mcp

Claude Code detects our auth challenge, opens the consent screen on pingaura.ai in your browser, you sign in and approve, and the CLI receives the token. Tokens refresh automatically.

With an API key:

claude mcp add pingaura --transport http https://www.pingaura.ai/api/mcp \
  --header "Authorization: Key YOUR_API_KEY"

Verify either way with:

claude mcp list

No restart required.

Claude Desktop

With OAuth: in Claude Desktop, go to Settings → Connectors → Add custom connector and paste https://www.pingaura.ai/api/mcp. Approve in the browser when it opens. No config file edit needed.

With an API key, edit the config file:

• macOS: ~/Library/Application Support/Claude/claude_desktop_config.json
• Windows: %APPDATA%\Claude\claude_desktop_config.json

{
  "mcpServers": {
    "pingaura": {
      "url": "https://www.pingaura.ai/api/mcp",
      "headers": {
        "Authorization": "Key YOUR_API_KEY"
      }
    }
  }
}

Restart Claude Desktop after saving.

Cursor

With OAuth (Cursor 0.45+): in Settings → Cursor Settings → MCP, click Add new MCP server, paste https://www.pingaura.ai/api/mcp, leave headers empty, and approve in the browser when prompted.

With an API key, add to ~/.cursor/mcp.json (global) or .cursor/mcp.json (project):

{
  "mcpServers": {
    "pingaura": {
      "url": "https://www.pingaura.ai/api/mcp",
      "headers": {
        "Authorization": "Key YOUR_API_KEY"
      }
    }
  }
}

Or go to Settings > Cursor Settings > MCP and add the server there. No restart required.

Codex

Codex supports MCP servers in the CLI and IDE extension.

Use the PingAura MCP endpoint:

• Server URL: https://www.pingaura.ai/api/mcp
• Authorization header: Key YOUR_API_KEY

For a step-by-step guide, follow OpenAI's Codex MCP setup. Use the PingAura server URL and API key shown above.

VS Code (GitHub Copilot)

Add to .vscode/mcp.json in your project:

{
  "inputs": [
    {
      "type": "promptString",
      "id": "pingaura-api-key",
      "description": "PingAura API Key",
      "password": true
    }
  ],
  "servers": {
    "pingaura": {
      "type": "http",
      "url": "https://www.pingaura.ai/api/mcp",
      "headers": {
        "Authorization": "Key ${input:pingaura-api-key}"
      }
    }
  }
}

VS Code will prompt you for the API key on first use.

Windsurf

Edit ~/.codeium/windsurf/mcp_config.json:

{
  "mcpServers": {
    "pingaura": {
      "serverUrl": "https://www.pingaura.ai/api/mcp",
      "headers": {
        "Authorization": "Key YOUR_API_KEY"
      }
    }
  }
}

Note: Windsurf uses serverUrl instead of url.

Amazon Q Developer

Add to ~/.aws/amazonq/mcp.json:

{
  "mcpServers": {
    "pingaura": {
      "type": "http",
      "url": "https://www.pingaura.ai/api/mcp",
      "headers": {
        "Authorization": "Key YOUR_API_KEY"
      }
    }
  }
}

Perplexity

Perplexity supports custom remote MCP connectors on paid plans.

1. Open the Perplexity app and go to Account Settings > Connectors.
2. Click + Custom connector and select Remote.
3. Fill in the details:
   • Display name: PingAura
   • MCP Server URL: https://www.pingaura.ai/api/mcp
   • Description: AI visibility scores, site audits, articles, and competitor data from PingAura
   • Authentication: Select API Key and enter your PingAura API key
   • Transport: Select Streamable HTTP
4. Check the acknowledgement box and click Add.
5. Click the PingAura connector card to enable it.

Once enabled, ask Perplexity questions like "What's my latest visibility score?" It will query PingAura directly.

ChatGPT

ChatGPT supports remote MCP servers through Developer Mode. This is available on Plus, Pro, and Team plans.

1. Open ChatGPT on the web or desktop app.
2. Open a new chat and enable Developer Mode from the model picker.
3. Click the Tools icon (wrench) in the composer, then Add MCP connector.
4. Enter the details:
   • Name: PingAura
   • URL: https://www.pingaura.ai/api/mcp
   • Authentication: Select API Key and enter your PingAura API key
5. Click Add to save.

Once connected, ChatGPT can call PingAura tools in Developer Mode chats.

Gemini CLI

Edit ~/.gemini/settings.json (global) or .gemini/settings.json (project):

{
  "mcpServers": {
    "pingaura": {
      "httpUrl": "https://www.pingaura.ai/api/mcp",
      "headers": {
        "Authorization": "Key YOUR_API_KEY"
      }
    }
  }
}

Note: Gemini CLI uses httpUrl instead of url.

Cline (VS Code extension)

Cline is a VS Code extension for AI-assisted coding. Its MCP settings live in a per-machine JSON file.

Edit ~/Library/Application Support/Code/User/globalStorage/saoudrizwan.claude-dev/settings/cline_mcp_settings.json (macOS) or the Windows equivalent under %APPDATA%\Code\User\globalStorage\saoudrizwan.claude-dev\settings\:

{
  "mcpServers": {
    "pingaura": {
      "url": "https://www.pingaura.ai/api/mcp",
      "headers": {
        "Authorization": "Key YOUR_API_KEY"
      },
      "disabled": false,
      "autoApprove": []
    }
  }
}

Reload VS Code after saving. Cline picks up the new server automatically.

AnythingLLM (self-hosted)

AnythingLLM is a self-hosted AI platform for building RAG pipelines and running agents. Add PingAura under the MCP server settings.

1. Open Settings → Tools → MCP Servers.
2. Click Add MCP Server.
3. Fill in:
   • Name: PingAura
   • Type: Streamable HTTP
   • URL: https://www.pingaura.ai/api/mcp
4. Under Headers, add Authorization: Key YOUR_API_KEY.
5. Click Save and enable the server.

PingAura tools are available to agents and custom AI workflows immediately.

Continue (open-source)

Continue is an open-source AI coding assistant that works inside VS Code and JetBrains. Add PingAura to ~/.continue/config.json:

{
  "mcpServers": [
    {
      "name": "pingaura",
      "transport": {
        "type": "streamable-http",
        "url": "https://www.pingaura.ai/api/mcp",
        "headers": {
          "Authorization": "Key YOUR_API_KEY"
        }
      }
    }
  ]
}

No restart required — Continue detects config changes and reconnects.

Zed (editor)

Zed supports MCP context servers via ~/.config/zed/settings.json. Add a context_servers entry:

{
  "context_servers": {
    "pingaura": {
      "settings": {
        "url": "https://www.pingaura.ai/api/mcp",
        "headers": {
          "Authorization": "Key YOUR_API_KEY"
        }
      }
    }
  }
}

Restart Zed after saving. The PingAura tools will appear in the Assistant panel.

Goose (by Block)

Goose is an open-source AI agent CLI from Block. Add PingAura as an extension in ~/.config/goose/config.yaml:

extensions:
  pingaura:
    type: streamable_http
    url: https://www.pingaura.ai/api/mcp
    headers:
      Authorization: "Key YOUR_API_KEY"
    enabled: true

No restart needed — run goose and PingAura tools are available in the session.

JetBrains AI Assistant (IntelliJ / PyCharm / WebStorm)

JetBrains AI Assistant supports remote MCP servers through the IDE settings UI.

1. Open Settings (⌘, on macOS, Ctrl+Alt+S on Windows/Linux).
2. Navigate to Tools → AI Assistant → MCP Servers.
3. Click + to add a new server and fill in:
   • Name: PingAura
   • URL: https://www.pingaura.ai/api/mcp
   • Headers: add Authorization → Key YOUR_API_KEY
4. Click Apply and OK.

The PingAura tools will be available in AI Assistant chat immediately — no IDE restart required.

Cherry Studio (desktop)

Cherry Studio is a desktop AI client with a built-in MCP server manager.

1. Open Settings → MCP Servers.
2. Click Add Server.
3. Fill in:
   • Name: PingAura
   • Type: Streamable HTTP
   • URL: https://www.pingaura.ai/api/mcp
4. Under Headers, add Authorization: Key YOUR_API_KEY.
5. Click Save and toggle the server on.

PingAura tools are ready in new conversations immediately.

5ire (desktop)

5ire is a desktop AI assistant with a tools panel for MCP integration.

1. Open the Tools section from the left sidebar.
2. Click New to add an MCP server.
3. Fill in:
   • Name: PingAura
   • URL: https://www.pingaura.ai/api/mcp
4. Add a custom header: Authorization: Key YOUR_API_KEY.
5. Save and enable the server.

The PingAura tools appear in the tools list and are available in chat.

Open WebUI (self-hosted)

Open WebUI is a self-hosted ChatGPT-style interface that supports MCP tool servers. Add PingAura under the Tools settings.

1. Open Settings → Tools → MCP Servers.
2. Click Add Server and fill in:
   • Name: PingAura
   • URL: https://www.pingaura.ai/api/mcp
3. Add a custom header: Authorization: Key YOUR_API_KEY.
4. Save the server configuration.
5. In your model settings, make sure Function Calling is enabled — Open WebUI only routes tool calls when the active model has function calling turned on.

PingAura tools will be available in new conversations once function calling is active.

Sourcegraph Cody (VS Code / JetBrains)

Cody is Sourcegraph's AI coding assistant. It supports remote MCP servers through the Cody settings panel.

1. Open the Cody sidebar in VS Code or your JetBrains IDE.
2. Navigate to Cody Settings → MCP Servers.
3. Click Add Server and fill in:
   • Name: PingAura
   • URL: https://www.pingaura.ai/api/mcp
4. Add a custom header: Authorization: Key YOUR_API_KEY.
5. Save. Cody picks up the new server without a restart.

PingAura tools are available in Cody chat and inline edit sessions.

Tome (by Runebook)

Tome is an MCP-first desktop AI client. Adding MCP servers is a first-class workflow.

1. Open the Tome app and go to Settings → Integrations.
2. Click Add MCP Server.
3. Fill in:
   • Name: PingAura
   • URL: https://www.pingaura.ai/api/mcp
4. Add a custom header: Authorization: Key YOUR_API_KEY.
5. Save. Tome connects immediately — no restart required.

LibreChat (open-source)

LibreChat is a self-hosted ChatGPT alternative. Add PingAura to your librechat.yaml config file under the mcpServers key:

mcpServers:
  pingaura:
    type: streamable-http
    url: https://www.pingaura.ai/api/mcp
    headers:
      Authorization: "Key YOUR_API_KEY"

Restart your LibreChat instance after saving. PingAura tools will be available to all configured AI endpoints.

Other MCP clients

The following clients also support PingAura via the standard streamable HTTP transport. Use the same URL (https://www.pingaura.ai/api/mcp) and Authorization header pattern shown above — refer to each client's documentation for where to configure remote MCP servers:

• Roo Code — VS Code extension; supports MCP servers in its settings panel
• Witsy — desktop AI assistant; MCP configured in its Integrations settings
• Bolt.diy — open-source AI coding assistant; MCP configured in settings
• Msty — desktop AI client; add PingAura under Settings → MCP Servers
• Jan — offline-first desktop AI; add PingAura as a remote MCP tool in Extensions → MCP
• Pinokio — AI app browser; configure PingAura as an MCP server in its tool settings

Available tools

Once connected, your AI assistant has access to these tools:

Read tools

Write tools

Example prompts

Once PingAura is connected, try asking your AI assistant:

• "Show me all my domains and their current status"
• "What's the latest visibility score for example.com?"
• "List the critical audit issues for my homepage"
• "Show me all published articles and their word counts"
• "Compare my visibility scores against competitors"
• "Which pages have the most visitors?"
• "Get the AEO issues from the latest audit"
• "Archive the prompt about delivery services"
• "Add hubspot.com as a competitor for my domain"
• "Mark that broken-links issue as resolved"
• "Publish my latest article"
• "How much of my prompt quota have I used?"

Access control

• API keys are scoped to a team account. All data returned is isolated to that account.
• Only account owners can create, update, or revoke API keys.
• Team members can view existing keys but cannot manage them.
• Keys support optional expiration dates.
• Revoked keys are immediately rejected.

Troubleshooting

"Unauthorized" error

• Check that your API key is correct and not revoked.
• Ensure the header format is Authorization: Key pa_xxx_yyy or Authorization: Bearer pa_xxx_yyy.

Tools not showing up

• Run claude mcp list (Claude Code) or check MCP status in your client.
• Verify the URL is exactly https://www.pingaura.ai/api/mcp.

Connection timeout

• The MCP server is stateless — each request is independent. If you're behind a firewall, ensure outbound HTTPS to www.pingaura.ai is allowed.