Onboard MCP

Platform

Search

⌘K

Onboard MCP

MCP guides

MCPCustomer Success MCP & Customer Onboarding AIIntroduction to MCPMCP setupConnecting to Onboard MCPSecurity & permissions (MCP)MCP tools, prompts, and discoveryMCP use cases — AI agentsExample workflowsAutomation platforms & MCPOttie — ChatGPT / Codex pet

Desktop clients

Back to Platform

MCP

Model Context Protocol for Onboard — operational AI for customer success and onboarding, with hosted tools, prompts, and desktop client setup.

Operational AI on live Onboard data

MCP

Connect Claude, Amazon Quick, Cursor, or other MCP clients once—then ask questions, run briefs, and draft updates against the same projects and customers your team already manages in Onboard.

Model Context Protocol (MCP) lets AI assistants work inside your onboarding and CS operations—not as a separate chatbot, but as a copilot that reads projects, tasks, KPIs, discussions, and customer context with the same permissions as your API key or signed-in user.

Onboard hosts MCP at https://rest.onboard.io/mcp/…. Assistants discover intent-based tools, call them with natural language or structured arguments, and receive action-ready answers with IDs, confidence, and suggested next steps. Write tools default to preview; humans approve customer-facing changes.

For search-oriented overviews (Customer Success MCP, customer onboarding automation, client onboarding AI agents), see Customer Success MCP & Customer Onboarding AI.

What teams do with it

Start with one narrow job, prove value in a week, then add focused agents.

Morning stand-up

Ask “What projects need attention today?” Tools like list_active_projects, find_blocked_tasks, and summarize_onboarding_status return a short brief—no dashboard export.

QBR / exec prep

Pull KPI health and account risks with get_kpi_dashboard, get_customer_workspace_context, and identify_project_risks. Validate before sending externally.

Post-call follow-up

Map notes to tasks and drafts via recommend_next_steps, create_tasks (preview), and draft_customer_email. Apply comments or updates only after review (apply=true).

Launch week

Surface blockers and owners with get_project_context, find_blocked_tasks, and assign_owner for stand-ups and Friday rollups.

Guardrails

• Human in the loop — AI drafts; humans commit customer-facing changes in Onboard.
• Least data — Prefer IDs and tool-backed fields over pasting sensitive customer text into prompts.

What you get

Capability	Meaning
Workflow tools	Search customers, summarize status, list KPIs, find blockers, draft updates—without REST paths.
Preview-before-write	Creates, comments, and updates default to preview; set apply=true when ready.
Hosted MCP (HTTP/SSE)	Connect remote clients to https://rest.onboard.io/mcp/sse (MCP setup).
API key or OAuth	Most desktops use an API key; Claude custom connectors use per-user OAuth (Connecting).
[API] tools	Browse the API Reference, then read_api_request / write_api_request when needed.
Prompts & resources	Built-in playbooks and context schemas—see MCP tools, prompts, and discovery.

Product language: say Project to customers; OpenAPI still uses tag Map for filters (Introduction).

Rollout

1. Pick one client — Claude Desktop, Amazon Quick, or Cursor (all clients).
2. One API key or OAuth connector — Security & permissions (MCP).
3. One agent, one job — Expand after the first workflow sticks (MCP use cases — AI agents).
4. Human gates — Confirm in Onboard before customer-visible sends.

Use MCP alongside REST and webhooks for pipelines; see Automation platforms & MCP for Zapier/Make/n8n/Workato.

Learning path

1. Introduction to MCP — concepts and Projects vs Map
2. MCP setup — production URLs and API key
3. Connecting to Onboard MCP — OAuth, Inspector, Client ID
4. Security & permissions (MCP) — who gets access
5. MCP tools, prompts, and discovery — full catalog
6. MCP use cases — AI agents · Example workflows
7. MCP desktop clients — per-client guides

Also: Customer Success MCP hub · Ottie pet · Security checklist · API Reference

How is this guide?

GoodBad

Customer Success MCP & Customer Onboarding AIUse Onboard MCP for customer success, customer onboarding automation, and client onboarding AI agents—live projects, blockers, KPIs, and safe writes from Claude, ChatGPT, Cursor, and more.

On this page

What teams do with itGuardrailsWhat you getRolloutLearning path

{"@context":"https://schema.org","@id":"https://docs.onboard.io/#organization","@type":"Organization","name":"Onboard","url":"https://onboard.io/","logo":"https://docs.onboard.io/brand/onboard-logo-large.svg","sameAs":["https://github.com/viaonboard/onboard-platform","https://onboard.io/"]}
{"@context":"https://schema.org","@type":"SoftwareApplication","name":"Onboard","applicationCategory":"BusinessApplication","operatingSystem":"Web","description":"Onboard is a customer onboarding platform with REST APIs, webhooks, MCP integrations, and automation for implementation teams.","url":"https://onboard.io/","documentation":"https://docs.onboard.io/docs/welcome","publisher":{"@id":"https://docs.onboard.io/#organization"}}
{"@context":"https://schema.org","@type":"TechArticle","headline":"MCP","description":"Model Context Protocol for Onboard — operational AI for customer success and onboarding, with hosted tools, prompts, and desktop client setup.","author":{"@id":"https://docs.onboard.io/#organization"},"publisher":{"@id":"https://docs.onboard.io/#organization"},"mainEntityOfPage":{"@type":"WebPage","@id":"https://docs.onboard.io/docs/mcp"}}
{"@context":"https://schema.org","@type":"BreadcrumbList","itemListElement":[{"@type":"ListItem","position":1,"name":"Onboard documentation","item":"https://docs.onboard.io/docs/welcome"},{"@type":"ListItem","position":2,"name":"MCP","item":"https://docs.onboard.io/docs/mcp"}]}