MCP is included in the Professional Plan and can be trialed for 30 days, free of charge.
Teamflect's Model Context Protocol (MCP) integration allows you to use Teamflect directly from supported AI tools.
Once connected, you can ask your AI tool to retrieve supported Teamflect information and perform supported actions without constantly switching between applications.
For example:
"Show me my active Teamflect goals."
"What Teamflect reviews require my attention?"
"Show me my next one-on-one and its talking points."
"Create a Teamflect task for me to follow up on today's action items."
"Update my Teamflect goal progress to 75%."
This guide explains how to use Teamflect with Claude, Claude Code, and ChatGPT, and provides practical prompts to help you get started.
Before You Start
Make sure:
Your organization has enabled Teamflect MCP.
Your Teamflect admin has granted you MCP access.
You have an active Teamflect account.
You have access to the AI tool you want to use.
Important: Connecting Teamflect to an AI tool does not give you additional Teamflect permissions. Your existing Teamflect user access continues to apply.
Step 1: Find Your Teamflect MCP Server URL
Some AI tools require your Teamflect MCP Server URL during setup.
To find it:
Open Teamflect.
Go to My Settings.
Locate the Connection to AI Tools (MCP) section.
Copy your MCP Server URL.
The URL tells a supported AI tool where to connect to Teamflect. It does not give someone access to your Teamflect account by itself. Teamflect authentication is still required.
Using ChatGPT? If your organization has already made Teamflect available as a custom app in your ChatGPT workspace, you may not need to use your MCP Server URL yourself. Follow the ChatGPT instructions here.
Option 1: Connect Teamflect to Claude
The exact setup depends on whether you're using Claude through an organization or an individual Claude account.
If your organization uses Claude Team or Enterprise, your Claude Owner or Primary Owner must first make Teamflect available as a connector for your organization.
Once Teamflect has been added to your organization:
Step 1: Open Your Connectors
In Claude, go to: Customize > Connectors
Step 2: Find Teamflect
Locate Teamflect among the connectors available to your organization.
Select Connect (+).
Step 3: Authenticate with Teamflect
You'll be redirected to complete Teamflect authentication.
Sign in using your Teamflect account and authorize the connection.
Agree and allow access to your AI tool.
Once authentication is complete, return to Claude.
Step 4: Enable Teamflect in Your Conversation
Open a Claude conversation.
Use the + button and select Connectors to manage which connectors are available for the conversation.
Enable Teamflect.
Step 5: Test Your Connection
Ask:
"Who am I in Teamflect?"
Then try:
"Show me my active Teamflect goals."
If Claude can retrieve your Teamflect information, you're ready to go!
Add Teamflect to an Individual Claude Account
If you're using a supported individual Claude plan and your organization allows you to configure your own connectors:
Go to Customize > Connectors.
Select the + button.
Select Add custom connector.
Enter Teamflect as the connector name.
Paste your Teamflect MCP Server URL.
Select Continue.
Select Connect and complete Teamflect authentication.
Once connected, test it with:
"Who am I in Teamflect?"
Option 2: Connect Teamflect to Claude Code
If you use Claude Code, you can add Teamflect directly from your command line.
You'll need your MCP Server URL from: Teamflect > My Settings > MCP
Step 1: Add Teamflect
Run:
claude mcp add --transport http teamflect YOUR_TEAMFLECT_MCP_SERVER_URL
Replace
YOUR_TEAMFLECT_MCP_SERVER_URLwith your Teamflect MCP Server URL.
Step 2: Authenticate with Teamflect
Open Claude Code and enter:
/mcp
Locate Teamflect and follow the authentication instructions.
If browser authentication is required, complete the Teamflect sign-in and authorization process in your browser.
Then return to Claude Code.
Step 3: Verify Your Connection
Run:
claude mcp list
Teamflect should appear among your configured MCP servers.
Step 4: Test Teamflect
Try:
"Who am I in Teamflect?"
Then:
"Show me my active Teamflect goals."
You can now use supported Teamflect capabilities directly through Claude Code.
Option 3: Use Teamflect with ChatGPT
Teamflect works differently in ChatGPT. If your organization uses a supported ChatGPT workspace, your ChatGPT Admin or Owner may configure and publish Teamflect as a custom app for your workspace. This means you may not need to manually enter your Teamflect MCP Server URL in ChatGPT.
Don't see Teamflect in ChatGPT? Contact your ChatGPT workspace admin to confirm that the Teamflect app has been made available to you.
Step 1: Find Teamflect
Open ChatGPT on the web.
Go to: Settings > Apps (or Plugins, depending on what your account shows).
Look for Teamflect among the apps available in your workspace.
Step 2: Connect Your Teamflect Account
If Teamflect displays a Connect option, select it.
Complete the Teamflect authentication and authorization process.
Sign in using your Teamflect account when prompted.
Once complete, return to ChatGPT.
Good to know: Your existing Teamflect permissions continue to apply. Connecting Teamflect to ChatGPT does not provide additional Teamflect access.
Step 3: Use Teamflect in a Chat
Start a new ChatGPT conversation.
Select Teamflect from the apps/tools available for your message, or refer to Teamflect in your prompt where supported.
Then ask what you need.
For example:
"Show me my active goals in Teamflect."
"What Teamflect reviews require my attention?"
"Who are my direct reports in Teamflect?"
ChatGPT can use the relevant Teamflect tools available through the app.
Step 4: Take Actions in Teamflect
If your organization has made the relevant Teamflect actions available, you can also ask ChatGPT to make supported changes. For example:
"Create a Teamflect task for me to prepare the quarterly performance review presentation by October 15."
"Create a Teamflect goal to complete our customer onboarding redesign by December 15."
"Update my Customer Onboarding goal progress to 75%."
"Create a Teamflect recognition for Sarah for supporting the team during the product launch."
ChatGPT may ask you to confirm an action before making a change. Review important information such as the employee, goal, task owner, dates, progress values, and descriptions before confirming.
Step 5: Use Teamflect in Follow-Up Messages
Selecting an app in ChatGPT applies to the message where you use it. If a later message requires ChatGPT to retrieve new Teamflect information or perform another Teamflect action, select or mention Teamflect again.
For example:
"Use Teamflect to show me my active goals."
Then, if you need fresh data or another action:
"@Teamflect update my Customer Onboarding goal progress to 75%."
You don't need to reselect Teamflect simply to discuss information already returned earlier in the conversation.
What Can I Do with Teamflect MCP?
Teamflect MCP includes both read and write/modify tools.
Read tools retrieve supported Teamflect information.
Write and modify tools allow supported changes to be made in Teamflect.
You don't need to memorize the tool names. In most cases, simply describe what you want to accomplish.
Available Teamflect MCP Tools
1. Read Tools
Ask Teamflect Help
Find answers to questions about using Teamflect. Try: "How do I create a 360-degree feedback cycle in Teamflect?"
Find Me
Identifies your authenticated Teamflect account. Try: "Who am I in Teamflect?"
Find a User
Find a colleague in Teamflect. Try: "Find Alex Johnson in Teamflect."
Get One-on-One Meetings
Retrieve supported one-on-one meeting information, including agendas, talking points, and notes. Try: "Show me my next one-on-one with Alex and its talking points."
Get Calendar
Retrieve upcoming Teamflect events, meetings, and deadlines. Try: "What's coming up for me in Teamflect this week?"
Get Check-ins
Retrieve supported check-in records and submitted answers. Try: "Show me my recent Teamflect check-ins."
Get My Direct Reports
Retrieve your direct reports. Try: "Who are my direct reports in Teamflect?"
Get Feedback
Retrieve supported feedback information. Try: "Show me the feedback I've recently received in Teamflect."
Get Goals
Retrieve supported goals and progress information. Try: "Show me my active goals and summarize their current progress."
Get Individual Development Plan
Retrieve supported Individual Development Plan information. Try:"Show me my Individual Development Plan and summarize my current focus areas."
Get Nine Box
Retrieve supported nine-box information where your Teamflect access permits it.
Get Company Policies
Retrieve company policies stored in Teamflect. Try: "Find our remote working policy in Teamflect and summarize it."
Get Recognitions
Retrieve supported recognition information. Try:"Show me the recognitions I've received recently."
Get Reviews
Retrieve supported performance review information. Try: "Show me the Teamflect reviews that currently require my attention."
Get Role Profile
Retrieve supported role information. Try: "Show me the responsibilities and expected skills for my role."
Get Succession Plan
Retrieve supported succession planning information where your Teamflect access permits it.
Get Surveys
Retrieve supported surveys and results.
Get Tasks
Retrieve Teamflect tasks and their current status. Try: "Show me my open Teamflect tasks that are due this week."
2. Write and Modify Tools
Depending on the tools available to you, your AI assistant can also help you take action.
Create Goal
"Create a Teamflect goal to improve customer onboarding completion to 90% by December 31."
Update Goal Progress
"Update my Customer Onboarding goal progress to 75%."
Edit Goal
"Change the due date of my Customer Onboarding goal to December 31."
Close Goal
"Close my Customer Onboarding goal."
Reopen Goal
"Reopen my Customer Onboarding goal."
Edit Goal Measurement Type
"Change the measurement type for my Customer Onboarding goal to percentage."
Create Task
"Create a Teamflect task for me to prepare the quarterly performance review presentation by October 15."
Create Recognition
"Create a Teamflect recognition for Sarah for supporting the team during the customer launch."
Confirm Teamflect Draft: When your AI assistant prepares supported Teamflect content as a draft, this tool can be used to confirm and save the prepared information to Teamflect.
Tip: Always review important details before confirming an action that creates or modifies Teamflect information.
Try These Workflows
The most useful MCP prompts often combine multiple pieces of Teamflect context rather than asking for a single record.
Prepare for a One-on-One
"Show me my upcoming one-on-one with Alex, including its talking points. Then summarize Alex's active goals so I can prepare for our conversation."
Prepare for Your Own Review
"Show me my active Teamflect goals and recent feedback and summarize the key themes I should consider before completing my performance review."
Prepare as a Manager
"Show me my direct reports and summarize their active Teamflect goals so I can prepare for my upcoming check-ins."
Turn a Discussion into Action
"Create a Teamflect task for me to follow up with Alex about the development action we discussed by Friday."
Review Your Priorities
"Show me my active goals, open tasks, and upcoming Teamflect deadlines and summarize what needs my attention this week."
Review Development Progress
"Show me my Individual Development Plan and summarize my development focus areas and planned actions."
Recognize Someone
"Create a Teamflect recognition for Sarah for proactively supporting the launch and helping the team meet the deadline."
Find a Company Policy
"Find our parental leave policy in Teamflect and summarize the key points."
Get Better Results
You don't need to use technical MCP commands when talking to your AI assistant.
Use normal language and provide enough context.
Instead of:
"Create a goal."
Try:
"Create a Teamflect goal to increase customer onboarding completion to 90% by December 31."
Instead of:
"Show goals."
Try:
"Show me my active Teamflect goals and summarize which ones need the most attention."
Instead of:
"Create recognition."
Try:
"Create a Teamflect recognition for Sarah for taking ownership of the customer launch issue and helping the team resolve it before the deadline."
Mentioning Teamflect explicitly can also help when your AI tool has access to several different applications.
How Do My Permissions Work?
Your AI tool interacts with Teamflect on your behalf.
Connecting Teamflect does not change what you're allowed to access.
If you can't access particular information in Teamflect, connecting an AI tool does not bypass that restriction.
Similarly, your AI tool can only perform Teamflect actions that are:
Supported by Teamflect MCP
Available through your AI platform
Allowed by your organization's configuration
Permitted by your Teamflect access
Troubleshooting
I Don't See MCP in Teamflect
Go to: Teamflect > My Settings
If MCP isn't available, contact your Teamflect admin to confirm that MCP is enabled and you've been granted access.
I Don't See Teamflect in Claude
If your organization uses Claude Team or Enterprise, your Claude Owner or Primary Owner may need to add Teamflect to the organization's connectors first. Once it has been added, check: Customize > Connectors
Claude Shows Teamflect but It Isn't Connected
Go to: Customize > Connectors
Locate Teamflect and select Connect.
Complete Teamflect authentication.
If authentication previously failed, disconnect and reconnect the connector.
Claude Code Can't Access Teamflect
First verify that Teamflect has been added:
claude mcp list
Then enter:
/mcp
Check the Teamflect connection and complete authentication if required.
I Don't See Teamflect in ChatGPT
Your ChatGPT workspace admin may need to publish the Teamflect custom app and make it available to you. Contact your ChatGPT admin if Teamflect doesn't appear in your available Apps.
ChatGPT Used Teamflect Earlier but Isn't Using It for My Follow-Up
ChatGPT app selection applies to the message where the app is used. If your new request requires fresh Teamflect data or another action, select Teamflect again or mention it in your request. For example: "@Teamflect show me the latest progress for my active goals."
My AI Tool Can't Find Information I Can See in Teamflect
The information may not currently be supported by an available Teamflect MCP tool.
Try a request related to one of the supported tools listed above.
I Can Retrieve Information but Can't Make a Change
Read and write capabilities are separate. The action may not be supported by Teamflect MCP or may not be enabled by your AI platform or organization.
Frequently Asked Questions
Do I need technical knowledge to use Teamflect MCP?
No. Once Teamflect is connected, you can interact with it using natural language.
Does my AI tool get access to everything in Teamflect?
No. Your existing Teamflect permissions continue to apply.
Can I use MCP to access information about my team?
Where supported, managers and admins can retrieve information their existing Teamflect permissions allow them to access.
Can my AI assistant make changes in Teamflect?
Yes, for supported and permitted write or modify actions.
Will I be asked before an AI tool changes something?
This depends partly on the AI platform and the action being performed. Review confirmation prompts carefully and verify important details before approving changes.
Why does setup look different between Claude and ChatGPT?
Each AI platform implements MCP differently. Claude uses connectors, while ChatGPT exposes MCP integrations as apps within supported workspaces. The underlying Teamflect permissions remain the same regardless of which supported AI tool you use.
Can Teamflect MCP capabilities change?
Yes. Additional tools and capabilities may be introduced over time.






