Overview

The Paylocity MCP Server enables AI platforms and developer tools to securely access and interact with Paylocity APIs and documentation using natural language. This allows partner and client developer teams to discover APIs, understand workflows, retrieve relevant documentation, and accelerate integration building directly from AI assistants such as Claude, ChatGPT, or Copilot.

🛑
This MCP is not for dynamic execution of tasks by AI agents
The MCP Server for API Documentation is intended as a tool to support engineering teams with discovery of what Paylocity APIs to leverage to implement a specific integration or workflow. AI-generated responses should be treated as informational guidance and not as a definitive source of truth. Always validate.
Do not share API keys, tokens, credentials, or any sensitive information with AI agents, prompts, or external tools connecting to this server.
AI tools may log or transmit inputs beyond your control, risking exposure of sensitive data. Only use credentials in secure, trusted environments. Never include them in prompts, queries, or AI workflows. Misuse can lead to unauthorized access, data leaks, and compliance violations.

Responsible Use & Security

Before connecting to the Paylocity MCP Server, review the following recommendations. These practices are intended to reduce security, privacy, and data-handling risks when working with third-party AI tools.

🔒
Core Security Principles

Treat all AI inputs as potentially logged or externally processed.

Never include credentials, tokens, or sensitive data in prompts, queries, or tool inputs.

Only operate within trusted environments and approved AI clients.

Use Reputable and Trusted MCP Clients Only

Connect to the Paylocity MCP Server only through applications that your organization trusts and permits. Examples of commonly used MCP-compatible applications include Claude, ChatGPT, Gemini, and Cursor.

Before using any AI application, review its security controls, privacy practices, data-retention policies, and organizational approval status. Unapproved or unfamiliar applications may create additional risks for confidential or sensitive information.

Keep Activity in a Separate Session

Use a dedicated conversation or workspace when accessing the Paylocity MCP Server. Where practical, do not enable unrelated MCP servers or connectors in the same session or allowing the agent to reference sites outside the paylocity.com domain for API instruction, as this can introduce unintended data sharing, allows search & source poisoning, or unpredictable behavior.

Review AI Platform Data-Use and Retention Settings

Prompts, retrieved documentation, and generated responses may contain proprietary, confidential, or otherwise sensitive business information.

Review the privacy and data-control settings available in your AI platform. Where supported and required by your organization, disable the use of conversations for model improvement and limit unnecessary storage or sharing of conversation data.

Depending on the platform, these controls may be identified by names such as:

Chat History & Training
Improve the Model
Data Controls
Training and Data Sharing
similar privacy or retention settings

The available controls and their names vary by provider and subscription type. Users are responsible for confirming that their selected settings comply with organizational security and data-handling requirements.

Verify AI Outputs

📘
Third-Party AI Disclaimer
AI applications and language models are operated independently of Paylocity. Their responses may be incorrect, incomplete, outdated, or inconsistent with current Paylocity documentation.
Users are responsible for validating generated information and reviewing proposed actions before relying on, approving, or executing them. Paylocity is not responsible for decisions, configuration changes, operational actions, or other outcomes resulting from reliance on content produced by a third-party AI service.
References to third-party products are included solely to explain possible connection methods. Such references do not imply that Paylocity is affiliated with, endorsed by, sponsored by, or officially partnered with the named providers unless separately stated by Paylocity.

AI outputs are not a source of truth and should be treated as a starting point. AI-generated tool requests should not be executed automatically. Keep automatic tool approval disabled unless your organization has explicitly authorized its use.

Executing AI generated code or relying on AI generated outputs, verify:

that suggested tool calls are appropriate
that technical guidance agrees with current Paylocity documentation
which Paylocity resources the AI application is attempting to access
what information will be submitted or retrieved
whether the generated request matches your intent
whether you are authorized to access the requested information

Development teams are responsible for testing all generated code and workflows before use. These manual reviews will help prevent unintended requests caused by unclear prompts or incorrect model interpretation.

Configuring AI Agents to use MCP

The Paylocity MCP Server can be connected to supported AI platforms (such as ChatGPT, Claude, or other MCP-compatible clients) to enable natural language discovery of Paylocity APIs and documentation. Once configured, developers can begin querying Paylocity APIs using natural language prompts to identify endpoints, understand workflows, and retrieve relevant documentation.

Below are some quick start steps for popular AI agents. It is recommended to refer to the platform documentation for connector configuration requirements and for any troubleshooting.

Connecting with ChatGPT

Requirements

ChatGPT subscription enabling MCP Agent capabilities

Configuration Steps

Open Apps from the ChatGPT navigation menu.
Select the Settings icon in the upper-right corner.
Under Enabled Apps, open Advanced Settings.
Review the developer-mode information and enable Developer mode.
Select Create app.
Enter Paylocity API Documentation in the Name field.
Under Connection, select Server URL and enter:

https://developer.paylocity.com/mcp?project=integrations
For Authentication, select No Auth.
Review and accept the displayed risk acknowledgment.
Optionally provide an app description and icon.
Click Create.
Select Connect.
Start a new ChatGPT conversation.
Click the + button and open More.
Select Paylocity API Documentation if it is not already active.

ChatGPT is now ready to begin referencing Paylocity API documentation for any prompts.

Connecting with Claude

Requirements

Claude subscription enabling MCP Agent capabilities

Configuration Steps

Open Claude and select Customize from the left navigation.
Select Connectors.
Click the + button.
Choose Add Custom Connector.
Enter Paylocity API Documentation as the connector name.
In the Remote MCP server URL field, enter:
https://developer.paylocity.com/mcp?project=integrations
Leave the advanced configuration fields empty.
Click Add.
Start a new Claude conversation.
Open + (Add files, connectors, and more) and select Connectors.
Confirm that Paylocity API Documentation is enabled.

Claude is now ready to begin referencing Paylocity API documentation for any prompts.

Connecting with Other AI Platforms

Requirements

Platform Supports acting as a MCP client, connecting to remote 3rd-party servers
Additional prerequisites such as subscription level to access features are variable depending on the platform
Before connecting, confirm that the application is approved for use by your organization and provides appropriate privacy, security, and tool-approval controls.

To connect from other platforms that support acting as an MCP client, follow their guidance documentation on connecting while using the Paylocity MCP Server URL:

https://developer.paylocity.com/mcp?project=integrations