Assembly MCP (Beta)
Connect Assembly to your AI assistant to bring recognition data into the tools your team already uses.
Last Updated: August 31, 2026
In this article:
This feature is currently in beta.
If you're interested in participating, contact us to request access.
Overview
Assembly already utilizes AI to provide meaningful insights across the platform. Assembly MCP extends this capability by giving Claude, ChatGPT, and other compatible AI experiences secure access to your Assembly data and recognition tools.
Assembly MCP can combine this with the other work tools your organization already has connected, such as Slack, Asana, or Microsoft Office, giving admins and managers more context to identify valued contributions, draft personalized recognition, and gain insights into team recognition and health.
Assembly MCP relies on two things: MCP capabilities, which let AI assistants securely retrieve information from Assembly or create recognition in it, and Assembly skills, which teach Claude how to do that well.
What is Assembly MCP?
Assembly MCP connects AI assistants with Assembly, providing a secure way to look up recognition information and create recognition on your behalf. Claude, ChatGPT, and other models can also draw on other work tools your organization has separately connected, such as Slack, Microsoft Office, or Salesforce, to bring in additional context alongside Assembly data.
With this additional context, AI assistants can help teams identify work that may otherwise go unnoticed and turn those contributions into more meaningful recognition.
Assembly MCP uses the Model Context Protocol (MCP) to provide Claude, ChatGPT, and others with secure, structured access to Assembly. It's beginner-friendly in nature, and ready to use out of the box.
What can I do with Assembly MCP?
Managers and admins can directly ask AI assistants questions about their teams and use the information available through their connected tools to:
- Identify meaningful contributions from direct reports
- Uncover meaningful work not already surfaced in Assembly
- Draft personalized recognition posts
- Identify team members who may need additional recognition or support
- Look for patterns related to recognition and team health
For example, you could ask Claude or ChatGPT:
"Summarize the most meaningful contributions from my direct reports this week and draft a recognition post for each person."
Claude or ChatGPT can use the available context to surface relevant accomplishments and create recognition drafts. Depending on the customization, drafts could include suggested point values and company values.
You can then review and edit the suggested recognition before choosing whether to post it to Assembly – no action is taken without your explicit approval.
How does it work?
Assembly MCP builds on Assembly's existing AI capabilities by giving AI assistants secure access to Assembly. The AI model then combines that with context from the other work tools you've already connected and uses Assembly skills to understand how to turn that information into thoughtful, on-brand recognition.
- Set up the Assembly MCP
- Assembly MCP is currently in beta testing. To gain early access, contact us here.
- Ask questions about your team:
Ask questions about your team's accomplishments, recognition, or team health. Here are several sample prompts to get you started:
Create personalized recognition
“Summarize the most meaningful contributions from my direct reports this week and draft a personalized recognition post for each person. Include a suggested point value and company value.”
Identify recognition gaps
“Which members of my team have received the least recognition recently? Look at their recent activity and identify opportunities to recognize their contributions.”
Prepare for a team meeting
“Give me a summary of the accomplishments and contributions from my team over the past month. Highlight specific examples I can celebrate in our next team meeting.”
- Review the results
AI assistants use the information available to you through your connected tools and Assembly to provide relevant insights or draft recognition. No action is taken without human approval. - Edit and post
Review the suggested recognition, make any changes you'd like, and choose whether to post it to Assembly. Recognition is not posted without someone manually choosing to do so.
FAQs
What is Assembly MCP?
Assembly MCP allows Claude, ChatGPT and other AI assistants to use relevant context from the work tools your organization already uses—such as Slack, Microsoft Teams, Outlook, Gmail, Asana, Linear, GitHub, Atlassian, Salesforce, and HubSpot.
This gives managers more context when identifying employee accomplishments and creating recognition, so meaningful contributions are less likely to go unnoticed.
Assembly MCP uses MCP (Model Context Protocol), an open standard that allows AI assistants to securely connect with external tools.
How is Assembly MCP different from Assembly's existing AI assistant?
Assembly's existing Dora AI helps you create recognition using context from within Assembly.
Assembly MCP extends this experience into AI assistants, allowing Claude, ChatGPT, and others to use relevant information from your connected work tools in addition to Assembly data.
Who can use Assembly MCP?
Assembly MCP is currently available through an early-access program for select organizations that:
- Have AI assistant access with an enterprise account
- Use Assembly
- Are participating in the Assembly MCP beta program
Is Assembly MCP required?
No. Assembly MCP is optional and does not replace Assembly's existing Dora AI assistant.
Does Assembly MCP cost extra?
There is no additional Assembly fee for Assembly MCP. However, your organization must have an external AI assistant solution, such as Claude Enterprise.
Which tools can AI assistants use with Assembly MCP?
You can connect work tools that support MCP and that your organization allows you to use with Claude, ChatGPT, and others.
Examples include Asana, Salesforce, Microsoft Office, Slack, Atlassian, GitHub, etc.
The tools available to you depend on your organization's AI assistant configuration. Contact your organization's admin for questions about available connections.
How does Assembly MCP help with recognition?
AI assistants can search and summarize relevant information about employee accomplishments across your connected work tools and use that context to draft recognition in Claude.
For example, instead of relying only on what a manager remembers or manually enters, Claude or ChatGPT can use relevant work activity from connected tools to help identify accomplishments and create a more informed recognition message.
Is recognition posted automatically?
No. A person must review and approve recognition before it is published.
For example, a manager can review the suggested recognition message, point value, and company value, make any edits they want, and then choose whether to post it to Assembly.
No recognition is published without direct user approval.
Interested in trying it out?
Assembly MCP is currently available through an early-access program for select organizations using Claude Enterprise.
If you're interested in participating in early access, reach out to learn more.