Bring Buddy into AI agents and assistants.
Give compatible AI clients a controlled way to analyze audiences, draft campaigns, build automations, and perform approved Buddy actions.
Buddy MCP is planned for future platform expansion and is not a general-availability product surface yet.
Today
Agent prepared a segment
Confirm before creating the audience or drafting the WhatsNew.
9:41 AM
Approved
9:41 AM
- Teams already working inside AI clients
- Developers building agent workflows around customer engagement
- Businesses that need strict approval before high-impact actions
Future platform
Let approved AI clients work with Buddy safely.
Buddy MCP is planned as a controlled interface for AI agents and assistants. The goal is to let trusted clients read useful context, prepare work, and request approved actions while preserving workspace permissions, confirmations, and audit logs.
Buddy MCP is planned for future platform expansion and is not a general-availability product surface yet.
Future platform
Bring Buddy into agent workflows with guardrails.
Buddy MCP is planned as a controlled way for compatible AI clients to read context, analyze audiences, draft campaigns, prepare automations and request approved actions. It is about permissions, confirmation and auditability, not uncontrolled automation.
Built for
Buddy MCP
Future platform
Read tools
Compatible clients may inspect audiences, engagement, subscriber context, campaign drafts, automation structures, and workspace settings.
Write tools
Approved clients may draft WhatsNews, create audiences, prepare automations, add tags, or update fields with confirmation.
Controlled actions
Sending messages, changing automation logic, exporting data, or making high-impact changes should require explicit confirmation.
Example workflow
Find customers who purchased in the last 30 days but have not purchased again, then draft a re-engagement WhatsNew for review.
Agent workflow
Authenticate.Inspect.Prepare.Confirm.
The useful version of AI agents in Buddy is narrow and reviewable: approved tools, workspace scopes, explicit confirmation for sensitive actions and a clear record of what happened.
Authenticate the client: Connect an AI client to a Buddy workspace with clear scopes and workspace-level permissions.
Read context: Allow the assistant to inspect approved audiences, engagement, subscribers, drafts, or automations.
Prepare work: Let the assistant draft a WhatsNew, suggest a segment, or prepare an automation structure.
Confirm sensitive actions: Require explicit approval for sends, exports, automation changes, and other high-impact operations.
Bring Buddy into agent workflows with guardrails.
Read tools
Compatible clients may inspect audiences, engagement, subscriber context, campaign drafts, automation structures, and workspace settings.
Write tools
Approved clients may draft WhatsNews, create audiences, prepare automations, add tags, or update fields with confirmation.
Controlled actions
Sending messages, changing automation logic, exporting data, or making high-impact changes should require explicit confirmation.
Future platform
Buddy MCP is planned as a future surface and should remain clearly labeled until it is ready.
Permissions first
AI clients should only access tools and data approved for the workspace.
Auditability
Teams should know what an assistant read, drafted, requested, and changed.
FAQ
Understand Buddy MCP.
Buddy MCP is planned as a controlled interface for AI agents and assistants. The goal is to let trusted clients read useful context, prepare work, and request approved actions while preserving workspace permissions, confirmations, and audit logs.
Still have questions?
Reach out to Sarah, the founder. She reads and replies to every message.
Message us