Apollo MCP

Apollo MCP lets compatible AI clients work directly with your Apollo account. You can search for prospects and companies, enrich records, create or update contacts, create sequences and variants, add contacts to sequences, and retrieve Apollo performance data from a natural-language conversation.

Apollo supports multiple ways to connect AI tools to Apollo MCP:

• Use Apollo's first-party connectors for AI platforms like ChatGPT, Claude, and Perplexity — no coding required.
• Use first-party coding plug-ins for apps like Codex or Coworker.
• Connect the standalone Apollo MCP server to any MCP-compatible client that supports remote servers over Streamable HTTP and OAuth.

Private Practice — Apollo prohibits AI model training with Apollo MCP integrations. Before you connect Apollo MCP to any AI client, confirm that model training is turned off in your AI account or client settings.

Endpoint

Use the following endpoint to connect to the standalone Apollo MCP server:

https://mcp.apollo.io/mcp

When you connect, your MCP client opens a browser-based OAuth flow. Sign in to Apollo and authorize access to your workspace, then you're ready to use Apollo with your AI tool.

Permissions and access

Apollo MCP reflects your Apollo permissions. The AI assistant can only perform actions that your Apollo user account, plan, and available credits allow.

For example:

• If you can't enrich contacts in Apollo, your AI assistant can't enrich contacts through MCP.
• If you don't have access to a sequence, your AI assistant can't add contacts to that sequence.
• If your workspace doesn't have enough credits for an enrichment action, your assistant can't complete the action.

Apollo plan limits, API limits, credit limits, and feature availability apply to MCP usage.

Before you start

Make sure you have:

• An active Apollo account.
• Permission to do the Apollo actions you want your AI client to perform.
• Available credits for enrichment or credit-consuming actions.
• An MCP-compatible client that supports remote MCP servers over Streamable HTTP with OAuth.
• A browser for the one-time OAuth sign-in.

Connect Apollo MCP

Add the Apollo MCP endpoint to your MCP-compatible client:

https://mcp.apollo.io/mcp

If your client asks for a transport type, choose Streamable HTTP. Some clients may label this as HTTP or Streaming HTTP.

After you save the server, complete the OAuth flow in your browser to connect your Apollo account.

Partners — If you're building an integration through a partnership with Apollo, register an OAuth app before connecting to Apollo MCP. For the full partner setup flow, check out Apollo Partnerships MCP Setup.

Client setup examples

The following examples show common MCP client configurations:

• Cursor
• Claude Code
• OpenAI Codex
• VSC with GitHub Copilot
• Claude Desktop
• Antigravity
• Other MCP clients

Cursor

Open Cursor settings and add a new MCP server.

Use the following configuration:

{
  "mcpServers": {
    "apollo": {
      "url": "https://mcp.apollo.io/mcp"
    }
  }
}

Save the configuration and restart Cursor. When you use an Apollo tool for the first time, complete the OAuth flow to connect your Apollo workspace.

Claude Code

Run the following command:

claude mcp add --transport http apollo https://mcp.apollo.io/mcp

Then run the following command in Claude Code and complete the OAuth flow:

/mcp

OpenAI Codex

Add Apollo MCP to your Codex configuration at ~/.codex/config.toml:

[mcp_servers.apollo]
url = "https://mcp.apollo.io/mcp"

Then authenticate:

codex mcp login apollo

Complete the OAuth flow to connect your Apollo account.

VS Code with GitHub Copilot

Create a .vscode/mcp.json file in your workspace:

{
  "servers": {
    "apollo": {
      "type": "http",
      "url": "https://mcp.apollo.io/mcp"
    }
  }
}

Open the Command Palette and run:

MCP: List Servers

Start the Apollo server and complete the OAuth flow when prompted.

Claude Desktop

Open Claude Desktop settings and go to Connectors.

Add a connector with the following URL:

https://mcp.apollo.io/mcp

Complete the OAuth flow to connect your Apollo workspace.

Antigravity

Follow the Antigravity instructions for connecting custom MCP servers and add the following to your mcp_config.json:

{
  "mcpServers": {
    "apollo": {
      "serverUrl": "https://mcp.apollo.io/mcp"
    }
  }
}

Save the configuration. Antigravity prompts you to complete the OAuth flow to connect your Apollo account.

Other MCP-compatible clients

For any other MCP-compatible client, add the Apollo MCP endpoint as a remote server:

https://mcp.apollo.io/mcp

Use Streamable HTTP as the transport and OAuth 2.0 for authentication.

Available actions

Most clients let you set each action to Always allow, Approval required, or Blocked. Apollo recommends Approval required for any action that consumes credits.

Missing an Action?

If an action isn't working or doesn't appear in your AI client, try disconnecting and reconnecting Apollo MCP.

Sample queries

Check out these sample queries for some MCP inspiration:

• Search and prospect
• Enrich people and companies
• Manage contacts
• Research companies
• Create and improve sequences
• Manage sequence enrollment
• Analyze performance

Search and prospect

Search for people, companies, contacts, and sequences using natural language.

Example prompts:

Find VP-level marketing leaders at SaaS companies with 200-1000 employees in the Bay Area.

Search for engineering managers at healthcare companies using Sumware Software.

Find Series B fintech companies in London with open sales roles.

People search returns profile and availability data. To retrieve email addresses, phone numbers, or other enriched details, use an enrichment action.

Enrich people and companies

Use enrichment tools to retrieve available contact and company data.

Example prompts:

Enrich this person: Jane Smith at Sumware Software.

Find the verified email for this contact using their name and company domain.

Enrich these 10 prospects and show me which ones have verified emails.

Enrichment actions consume credits based on your Apollo plan.

Manage contacts

Create and update Apollo contacts without switching tabs.

Example prompts:

Create a contact for John Doe, VP of Sales at Sumware Software.

Update this contact's title to Senior Director of Engineering.

Find the existing contact for this email address before creating a new one.

Research companies

Retrieve company details and hiring signals.

Example prompts:

Get company details for apollo.io, including employee count, industry, headquarters, and technologies.

Show me recent job postings for this company.

Find companies in the cybersecurity industry with 500-1000 employees.

Create and improve sequences

Create new sequences, revise existing messaging, and tailor outreach for specific personas, industries, or campaign goals.

Example prompts:

Create a 5-step outbound sequence for VP Marketing at mid-market SaaS companies. Keep the tone friendly and consultative. Make step 1 under 60 words and set the sequence inactive so I can review it before launch.

Review my Q2 Outbound sequence and suggest improvements to increase reply rates. Focus on subject lines, personalization, CTA clarity, and step spacing.

Rewrite this sequence for RevOps leaders at Series B SaaS companies. Keep the same offer, but make the messaging more concise and value-focused.

Create 3 alternate first-touch emails for this sequence. Each version should use a different angle: pain point, trigger event, and social proof.

Find sequences targeting VP Sales personas and show me which ones have the highest reply rate in the last 6 months.

Manage sequence enrollment

Use Apollo MCP to find sequences, list available sending mailboxes, and add or remove contacts from sequences.

Example prompts:

Show me my available email sequences.

List my connected email accounts.

Add these contacts to the Q2 Outbound sequence using my Gmail account.

Remove this contact from the active outbound sequence.

Before adding contacts to a sequence, list email accounts so you can choose the correct sending mailbox.

Analyze performance

Use analytics tools to retrieve available performance and engagement data.

Example prompts:

Which sequences had the highest reply rate in the last 6 months?

Show email engagement trends for this sequence.

Compare sequence performance by reply rate and meeting conversion.

Best practices

Be specific with filters. Include job titles, seniority, location, industry, company size, revenue range, and technologies when relevant.

Use company domains when enriching. A domain usually improves match accuracy compared to company name alone.

Search before enriching. People search doesn't return full contact details. Use enrichment only for the prospects you want to unlock.

Review credit-consuming actions before approving them. Most MCP clients let you require approval before specific tools run.

List email accounts before adding contacts to sequences. This helps you choose the correct mailbox and avoid enrolling contacts with the wrong sender.

Use deduplication when creating contacts. Ask the assistant to check for existing contacts before creating new records.

Pricing and credits

There is no additional cost to connect to Apollo MCP.

Your Apollo plan, permissions, and credits still apply.

Limitations

Apollo MCP has the following limitations:

• People search results don't include full contact details such as email addresses or phone numbers. Use enrichment to retrieve available contact details.
• Search results may be paginated. Ask your AI assistant to retrieve the next page when needed.
• Tool availability may vary by Apollo plan, workspace permissions, client support, and rollout status.
• Destructive actions, such as bulk delete, aren't supported to protect your data.

Privacy and security

Apollo MCP uses OAuth 2.0 authentication. Your MCP client doesn't need an Apollo API key for the standalone server setup.

Access is scoped to the Apollo user who authorizes the connection. The assistant can only access data and perform actions that the authorized Apollo user can access and perform.

You can revoke access from your connected app or account settings.

Before connecting Apollo MCP to an AI client, review that client's privacy and data controls. If your client offers model training or data retention settings, configure them according to your company's privacy and security requirements.

Troubleshooting