Skip to content

ATZ CRM MCP Server

The ATZ CRM MCP Server lets you connect ATZ CRM with AI assistants such as ChatGPT, Claude, Gemini, and other MCP-compatible clients.

Once connected, your AI assistant can securely work with your ATZ CRM data through approved tools. For example, you can ask it to find candidates, read job details, summarize notes, review matches, create records, or help prepare recruiter workflows without switching between multiple screens.

MCP stands for Model Context Protocol. It is an open standard that allows AI applications to connect with external systems, tools, and data sources.

In simple terms, MCP lets an AI assistant interact with ATZ CRM in a structured and secure way. Instead of manually copying data from ATZ CRM into an AI tool, the assistant can request the right information through MCP tools.

The ATZ CRM MCP Server is an integration that allows supported AI clients to connect with your ATZ CRM account.

It can help with recruiter workflows such as:

  • Searching candidates, jobs, organizations, contacts, matches, and notes
  • Reading full candidate, job, organization, and contact details
  • Creating or updating candidates, jobs, organizations, and contacts
  • Creating notes and activities
  • Creating candidate-job matches
  • Updating match stages
  • Reading metadata required for CRM workflows

The MCP Server uses your generated MCP URL as the credential. Do not share this URL publicly.

MCP Server Integration is available for ATZ CRM Enterprise Plan accounts.

The MCP Server Integration page is visible in Admin Settings, but only Enterprise Plan accounts can generate or rotate the MCP Server URL.

You need:

  • An active ATZ CRM Enterprise Plan account
  • Access to Admin Settings
  • Permission to manage integrations
  • A compatible MCP client such as ChatGPT, Claude, Gemini, or another MCP-compatible tool

Ask natural-language questions instead of manually searching through multiple CRM pages.

Example:

Find candidates with React experience in London.

Use your AI assistant to retrieve CRM information, summarize data, and prepare follow-up actions from one place.

Recruiters can quickly prepare candidate summaries, shortlist suitable profiles, review pipeline status, or create notes after calls.

The MCP Server works through defined tools and account-scoped access, so actions happen through controlled ATZ CRM workflows.

Your MCP Server URL is a credential.

Anyone with this URL may be able to connect an MCP client to your ATZ CRM account, depending on the enabled tools and account access. Treat it like a password.

Security recommendations:

  • Do not share the MCP URL in public channels
  • Do not paste it into documents or tickets unless required
  • Rotate the MCP URL if you think it has been exposed
  • Review tool calls in your AI client before approving actions
  • Disable MCP if your team is not actively using it

ATZ CRM stores only the hashed token. The raw MCP URL is shown only when generated or rotated.

How To Generate Your ATZ CRM MCP Server URL

Section titled “How To Generate Your ATZ CRM MCP Server URL”
  1. Log in to ATZ CRM.
  2. Go to Admin Settings.
  3. Open Integrations.
  4. Click MCP Server Integration.
  5. Review the security warning.
  6. Click Generate URL.
  7. Copy the generated MCP Server URL.

Important: The full URL is shown only once after generation or rotation. If you leave the page or refresh it, only a masked version will be visible.

If you need a new copyable URL later, use Rotate URL.

Use Rotate URL when:

  • The URL may have been shared with the wrong person
  • An employee with access has left the company
  • You want to reset access for connected AI clients
  • You need to connect a new MCP client and no longer have the full URL

Steps:

  1. Go to Admin Settings.
  2. Open MCP Server Integration.
  3. Click Rotate URL.
  4. Copy the new URL.
  5. Update your MCP client with the new URL.

Rotating the URL invalidates the previous credential.

  1. Go to Admin Settings.
  2. Open MCP Server Integration.
  3. Turn off MCP Server.
  4. Confirm the setting is disabled.

When disabled, MCP clients using the URL will not be able to connect.

Requirements may vary depending on your ChatGPT plan and OpenAI’s current connector availability.

General setup:

  1. Open ChatGPT settings.
  2. Go to Apps & Connectors.
  3. Open advanced connector settings if required.
  4. Create a custom MCP connector.
  5. Enter a name such as ATZ CRM.
  6. Paste your ATZ CRM MCP Server URL.
  7. For authentication, choose No authentication.

Why “No authentication”?

ATZ CRM uses the generated MCP URL itself as the credential. You do not need to add a Bearer token or API key.

After setup, start a chat and select the ATZ CRM connector when you want ChatGPT to use ATZ CRM tools.

Requirements may vary depending on your Claude plan and connector availability.

General setup:

  1. Open Claude settings.
  2. Go to Connectors.
  3. Add a custom connector.
  4. Enter a name such as ATZ CRM.
  5. Paste your ATZ CRM MCP Server URL.
  6. Save the connector.
  7. Start a conversation and allow Claude to use the connector when needed.

Most MCP clients ask before running tools. Review each tool request before approving it.

ATZ CRM currently supports the following MCP tool categories.

  • List candidates
  • Read candidate details
  • Create candidates
  • Update candidates

Example prompts:

Find candidates with C++ experience in London.

Open the full profile for this candidate.

Create a candidate named Priya Sharma with this email and phone number.

  • List jobs
  • Read job details
  • Create jobs
  • Update jobs

Example prompts:

Find active software engineering jobs.

Show me the job details for Senior Backend Developer.

Create a new job for Account Manager at Acme Corp.

  • List organizations
  • Read organization details
  • Create organizations
  • Update organizations

Example prompts:

Find organizations in the healthcare industry.

Show me all details for Acme Corp.

Create a new organization called Northstar Technologies.

  • List contacts
  • Read contact details
  • Create contacts
  • Update contacts

Example prompts:

Find contacts at Acme Corp.

Show me the hiring manager details for this organization.

Add a new contact for this client.

  • List matches
  • Create candidate-job matches
  • Update match stage

Example prompts:

Show candidates matched to the Senior Recruiter job.

Match this candidate to the Account Executive role.

Move this candidate to the interview stage.

  • List notes
  • Create notes

Example prompts:

Show notes for this candidate.

Add a note that the candidate is available for interviews next week.

  • Create activities

Example prompts:

Create a follow-up activity for tomorrow morning.

Schedule a call activity for this candidate.

  • List metadata used by MCP tools

Example prompts:

Show available metadata for creating a candidate.

What fields can I use for this workflow?

Prompt:

Find candidates with React experience in London. Show the strongest matches with their skills, location, current role, and recent notes.

Useful for:

  • Shortlisting candidates
  • Preparing outreach lists
  • Reviewing candidate fit quickly

Prompt:

I am interviewing this candidate tomorrow for the Senior Product Manager job. Pull the candidate profile, job details, and notes. Create a short interview prep brief with five questions.

Useful for:

  • Recruiter prep
  • Hiring manager briefings
  • Structured interviews

Prompt:

I have a meeting with Acme Corp. Find the organization, contacts, open jobs, and recent notes. Prepare a meeting brief with key talking points.

Useful for:

  • Account management
  • Client calls
  • Business development meetings

Prompt:

Show all candidates matched to the Software Engineer job. Group them by stage and summarize next actions.

Useful for:

  • Pipeline reviews
  • Recruiter standups
  • Hiring manager updates

Prompt:

Find the best candidates for the Full Stack Developer job. Read their profiles, compare them against the job, and suggest the top three candidates to match.

Useful for:

  • Faster shortlist creation
  • Urgent hiring needs
  • Candidate rediscovery

For best results, be specific in your prompt.

Good search examples:

Find candidates with Python and Django experience in Manchester.

Find contacts at organizations in the finance industry.

Show active jobs for Acme Corp.

Find candidates who mention Kubernetes, AWS, or DevOps.

You can search by common recruiting fields such as:

  • Name
  • Email
  • Phone
  • Location
  • Skills
  • Current role
  • Experience
  • Company
  • Job title
  • Organization name
  • Contact name
  • Match stage
  • Notes

Candidate search is optimized for richer candidate profile data, including skills and searchable profile text.

  • Use names, emails, job titles, and company names when possible.
  • Ask for a specific output format, such as a table, shortlist, email draft, or meeting brief.
  • Review tool calls before approving them in your MCP client.
  • Use follow-up prompts to refine results.
  • Rotate the MCP URL if access may have been exposed.
  • Disable MCP when it is not needed.

MCP URL generation is available only for Enterprise Plan accounts. If your account is not on Enterprise, you can view the page but cannot generate or rotate the URL.

Check that:

  • The MCP Server URL was copied correctly
  • MCP Server is enabled in ATZ CRM
  • The URL has not been rotated
  • Your AI client supports custom MCP connectors
  • You selected No authentication if the client asks for authentication

For security, ATZ CRM only shows the full URL once after generation or rotation. Rotate the URL to get a new copyable credential.

Try using more specific terms, such as skills, location, email address, job title, or organization name. For candidate searches, include important terms clearly in the prompt.

Example:

Find candidates with C++ experience in London.

ATZ CRM MCP Server follows the same request policy used by ATZ CRM Open API. Very large or repeated requests may be limited depending on your account and system protections.

List tools are paginated and capped to keep responses stable and fast.

AI assistants may summarize, transform, or interpret CRM data. Always review important outputs before sending them to candidates, clients, or hiring managers.

For write actions such as creating candidates, adding notes, or updating stages, review the tool call before approving it in your MCP client.

Reach out regarding any queries:)