MCP Server

Manatal MCP Server Overview

Connect your favorite AI tools to Manatal with our MCP Server. This allows your AI assistants to safely search, create, and manage your ATS data without constant tab-hopping.

What is MCP?

Model Context Protocol (MCP) is an open standard that connects AI applications and tools to external platforms like Manatal. With MCP, AI applications like ChatGPT and Claude can connect to data sources, tools, and workflows, allowing them to access additional information and perform tasks.

What is the Manatal MCP Server?

The Manatal MCP Server is our integration that directly connects with your AI tools to perform tasks like searching for candidates and jobs, creating notes, and managing matches. Any MCP-compliant client can connect to and use these capabilities, subject to your account's permissions.

Why use the Manatal MCP Server?

• Easy Setup: Connect with a simple URL authentication—no complex installation is required.

• Work from One Place: Ask your assistant to search, create, match, or add notes directly in Manatal.

• Faster Outcomes: Condense multi-step tasks into a single, natural-language request.

  MCP Server video guide

Who is it for?

• Recruiters: Ask for shortlists, summarize candidate profiles, and get suggested interview questions.
• Teams using AI copilots (like Claude or ChatGPT) for increased productivity.
• Developers and Partners: Build custom assistants or workflows that incorporate Manatal data.

What can you do with it?

• Candidates: Add new candidates, location, or status, and view full profiles.
• Jobs: Add new jobs, search or filter open roles, and view job details and pipelines.
• Organizations: Add organization (company/client) records and search for existing ones.
• Matches: Create matches between candidates and jobs, and search existing matches to review fit.
• Contacts: Search for contacts (e.g., hiring managers, referrers) to add context to your workflows.
• Notes: Add notes to candidates or jobs and view existing notes to keep collaboration in one place.

Connecting to ChatGPT

From Manatal:

1. Head to the following page or click on your photo at the top right-hand side of the screen, then "Settings" from the dropdown menu.
   
   
2. Click on “Manatal MCP”.
   
3. Identify the “ChatGPT” card and click on “Connect with ChatGPT”.
   
4. Once redirected to the ChatGPT Plugin Directory, click on "Install plugin" button.
   
5. You will be prompted to login with your Manatal Account (Google also supported).
   
6. Click on the"Connect with ChatGPT" button to provide your consent for ChatGPT to access Manatal.
   
7. Start interacting with Manatal directly from ChatGPT.
   
   

Alternatively, from ChatGPT:

1. In the ChatGPT plugin Directory, search for Manatal and click on the + icon next to it.
   
2. You will be prompted to login with your Manatal Account (Google also supported).
   
3. Click on the"Connect with ChatGPT" button to provide your consent for ChatGPT to access Manatal.
   
4. Start interacting with Manatal directly from ChatGPT.
   
   

Connecting to Other AI tools

This guide walks you through connecting your AI tools to Manatal using the Model Context Protocol (MCP).

You can connect to the Manatal MCP Server through the Integration Marketplace as shown below.

Supported Clients

Currently, the MCP Server is supported on the following AI tools:

• Claude.ai (Recommended for the best experience)
• Gemini CLI (For more technical users)
• Microsoft Copilot Studio
• Other MCP-supported clients (May be compatible, but we have not verified their behavior with our MCP Server)

Retrieving your MCP Server URL

1. Head to the following page by clicking on "Administration" from your side menu and open the "Integrations" category.
   
   
   Alternatively, head to the following page or click on your photo at the top right-hand side of the screen, then "Settings" from the dropdown menu and click on "Manatal MCP".
   
   
   

2. In the Integrations page, search for "Manatal MCP", and click “Enable” on the card.
   
   Alternatively, in the Manatal MCP Settings page, click on "Connect with other AI tools".
   

3. Review the integration details and disclaimers, accept the terms and conditions, and click “Integrate”.
   

4. You will now receive your MCP Server URL. Simply copy this URL to connect your AI tools.

Connect to Your AI Tool

Claude

1. Navigate to "Settings" in your Claude account.
2. Click on the “Connectors” section in the left-side panel.
3. Locate and click “Add custom connector”.
4. Paste the MCP Server URL from Manatal, give the connector a name (e.g., "Manatal ATS"), and click “Add”.
5. You can now access your Manatal database and tools inside Claude.

Gemini CLI

1. Install Gemini CLI

   1. Open your computer's Terminal (Mac/Linux) or Command Prompt (Windows).

   2. Type the following command and press Enter:

      1. npm install -g @google/gemini-cli
      Note

      This method requires that you have Node.js (https://nodejs.org/en/download/) installed on your computer first. If you don't have it, please install it before proceeding.

2. Authenticate your account

   • Option 1 (Recommended): Log in with your Google account.
   • Option 2: Use a Gemini API Key.
   • Option 3: Use Google Cloud's Vertex AI.

3. Connect to your Manatal MCP Server

   1. Find the settings.json file. It's usually located in a hidden folder named .gemini in your user's main directory.

   2. Open the settings.json file with a text editor (like Notepad, VS Code, or Sublime Text).

   3. Add the following section to the file. Make sure to replace the placeholder URL with your actual Manatal MCP Server URL:

      {
        "mcpServers": {
          "Manatal MCP": {
            "httpUrl": "<YOUR_MANATAL_MCP_SERVER_URL_HERE>"
          }
        }
      }
      
      

      Important

      Replace <YOUR_MANATAL_MCP_SERVER_URL_HERE> with the unique URL you received from Manatal.

Microsoft Copilot Studio

1. Open Microsoft Copilot Studio and open the agent you want to extend. If you need to create a new agent please refer to Microsoft’s documentation
2. Go to the agent’s Tools page.
3. Click “Add a Tool” → “New Tool”.
4. Select “Model Context Protocol” (this opens the MCP onboarding wizard).
5. Fill in:
   1. Server Name: Manatal MCP Server
   2. Server Description: Search and manage Manatal ATS data
   3. Server URL: Paste your Manatal MCP Server URL
   4. For Authentication type, select “None” (because the Manatal URL itself is the credential). Then click “Create”.
6. In the Add tool dialog:
   • Select “Create a new connection” (or choose an existing one), then click “Add and configure.”
7. Test the connection in chat (and connect if prompted)
   1. Open the agent’s Test chat (or the chat panel on the right).
   2. In the chat box, send a simple test prompt, for example:
      1. Search for candidates named "John"
   3. If Copilot Studio prompts you to verify credentials, click “Open connection manager.”
   4. In Connection manager, find the connection for the Manatal MCP Server you just created.
   5. Click “Connect”, then click “Submit.”
   6. Return to the chat and send the same test prompt again.

Microsoft documentation: For the latest Copilot Studio steps and UI labels, see Microsoft Learn’s guide to connecting to an MCP server.

Usage and Rate Limits

• Standard API rate limits apply to the Manatal MCP Server: 100 requests per minute.

Revoking Access

Revoke acess for all users

Admins can disconnect all users at once by turning off the Manatal MCP.

1. Head to the following page by clicking on "Administration" from your side menu and open the "Features" category.
   
   

2. Click on "Manatal AI".
   

3. Toggle off the Manatal MCP toggle.
   

4. Click "Continue".
   

Revoke access to individual tools

Users can revoke an individual tool's access to Manatal by selecting Revoke access on that tool's card. Other connected tools are unaffected.

1. Head to the following page or click on your photo at the top right-hand side of the screen, then "Settings" from the dropdown menu.
   
   
2. Click on “Manatal MCP”.
   
3. Click on "Revoke access" button on the specific tool's card.
   
4. Click "Continue".
   

Supported Tools and Example Use Cases

The Manatal MCP Server provides the following MCP tools. AI assistants use these tools to create, search, and manage content in your Manatal ATS. You can find a complete breakdown of the inputs and data structures for these interactions in this guide.

Common Use Cases (Tool Chaining Examples)

These tools work seamlessly together through prompts, and their real power comes from combining them. With a single prompt, you can search for top candidates, summarize them, match them to a job, and add a note. Understanding these tools will help you craft efficient prompts that tackle complex tasks by combining multiple tools.

Use Case 1: Interview Preparation from Candidate Profile & Job Description

Scenario: You have an upcoming interview and need to prepare relevant questions and talking points based on the candidate's background and the job requirements.

How to Use (Example Prompt):

I'm interviewing Daniel Wellington (daniel.wellington@email.com) tomorrow for our Senior Product Manager position for Acme Corp. Do the following:

1. Find the profile using their email address
2. Pull their complete candidate information
3. Check any notes our team has added about them
4. Find our "Senior Product Manager" job opening at Manatal
5. Get the complete job description and requirements
6. Review any notes about this role

Then synthesize all this information to create a one-page prep note (5 questions max) containing tailored interview questions that align their experience with our role requirements.

What You'll Get:

• Tailored interview questions that align the candidate's experience with job requirements.
• Key areas to probe during the conversation.
• Talking points about how their background fits the role.
• Specific examples from their history to discuss.

Use Case 2: Executive Summary Email for Hiring Manager

Scenario: Your hiring manager needs a quick overview of top candidates before making interview decisions.

How to Use (Example Prompt):

For the candidates selected below, prepare an executive summary for our client by doing the following:

1. Retrieve the full profiles of the following candidates:
- Daniel Wellington (daniel.wellington@email.com)
- Emma Johnson (emma.johnson@email.com)
- Martina Lopez (martina.lopez@email.com)

2. Then create a professional executive summary email with the following template:

Hi John,

Please find below the top candidates for your review. Thank you.

Candidate: (Candidate Display Name) (Candidate Location)
Candidate Highlights: (Summary of Candidate 2-3 Lines)

Work Experience:
- (Position 1) - (Company Name) - (Date Range) - (Tenure)
- (Position 2) - (Company Name) - (Date Range) - (Tenure)

What You'll Get:

• Professional, concise email ready to send.
• Key highlights of each candidate's strengths.
• Relevant experience and qualifications summarized.

Use Case 3: Comprehensive Candidate Pipeline Review

Scenario: You need to update stakeholders on the status of all active candidates across multiple positions.

How to Use (Example Prompt):

Create a pipeline status report for the Senior Data Scientist position at Acme Corp. Do the following:

1. Find the "Senior Data Scientist" job at "Acme Corp"
2. Get the job details and any notes about hiring progress
3. Find all candidates currently matched to this job
4. For each matched candidate, review their notes for status updates

Then compile a progress report with the following:
- Job overview: position title, status, and urgency
- Top 3 candidates: name, current stage, and brief status for each
- Next steps: what needs to happen to move candidates forward

What You'll Get:

• Overview of all candidates in your pipeline.
• Status of each candidate by hiring stage.
• Identification of any bottlenecks.
• Suggested next actions.
• Ready-to-share progress report.

Use Case 4: Client Relationship Management

Scenario: You're preparing for a call with a key client organization and need context on your relationship history and current engagements.

How to Use (Example Prompt):

I have a meeting with Acme Corp tomorrow. Help me prepare by doing the following:

1. Find "Acme Corp" in our client database
2. Find all contacts at "Acme Corp"
3. Find all active jobs at "Acme Corp"
3. Review all notes about our interactions with them

Then create a meeting brief with:
- Company overview: key details from our notes
- Key contacts: names and roles (max 3 people)
- Current jobs: position titles and status (max 3 positions)
- Talking points: 3 key items to discuss in the meeting

What You'll Get:

• Complete relationship history.
• Summary of previous conversations and commitments.
• Overview of current hiring needs.
• Talking points for your meeting.
• Context on key contacts and their preferences.

Use Case 5: Rapid Candidate Sourcing and Matching

Scenario: A new urgent job opening comes in and you need to quickly identify and match suitable candidates from your existing database.

How to Use (Example Prompt):

We just opened a Full Stack Developer role at Acme Corp. Help me find matches by doing the following:

1. Get the complete job details for the "Full Stack Developer" role at "Acme Corp"
2. Search for candidates with:
   - Current location in "San Francisco"
   - "Developer" in their current position
   - Limit to 20 results

3. For the candidates found:
   - Get their complete profiles & identify the top 3 candidates
   - Match them to the job

4. For each matched candidate, add a note explaining:
   "Matched to "Full Stack Developer" role. Fit reason: [brief explanation why]"

Then provide:
- Top 3 matched candidates with names and current companies
- Brief reason why each is a good fit (2-3 sentences)
- Suggested next step for outreach

What You'll Get:

• Ranked list of best-fit candidates.
• Rationale for each match.
• Gaps or concerns to address.
• Recommended outreach approach for each candidate.
• Quick start on filling urgent roles.

Tips for Best Results

1. Chain Related Actions: Ask for multiple things in one prompt - the AI can handle complex requests.
2. Use Human Identifiers: Use names, email addresses, job titles, and company names instead of just IDs.
3. Be Specific About Output: Tell the AI what format you want (email, report, bullet points, etc.).
4. Include Context: Mention why you're asking - it helps the AI prioritize information.
5. Add Details as You Go: You can always refine with follow-up questions.

Common Search Filters You Can Use

• By Company: "candidates who work at Google" or "current company is Amazon".
• By Position: "candidates with 'Engineer' in their current position".
• By Location: "candidates in San Francisco" or "address in New York".
• By Education: "candidates with degrees from MIT" or "latest university is Stanford".
• By Email/Phone: "candidate with email john@example.com".
• By Date: "candidates created after January 1, 2024" or "jobs updated in the last week".
• By Status: "active jobs" or "candidates in interview stage".