Q2 2026 Release Notes for Staircase AI

Create Automations Using the Staircase AI MCP Server

Users can generate personal OAuth 2.0 credentials (Client ID and Client Secret) to authenticate LLM clients with the Staircase AI MCP server. This enables secure, programmatic access to Staircase AI data without a browser login, making it easier to integrate automated clients and LLM tools like Claude.

This capability supports system-to-system connections, allowing customers to automate processes and enable machines, not just users, to interact with Staircase AI.

What’s New

• OAuth 2.0 Client Credentials Flow: Generate a Client ID and Client Secret from Profile settings and use them to obtain a short-lived JWT access token.
• Programmatic Access to MCP Server: Authenticate LLM clients to interact with the Staircase AI MCP server without manual login.
• Access to MCP Tools: Retrieve real-time customer insights using MCP tools, including account health summaries, risk signals, and stakeholder changes.
• User-Scoped Credentials: Credentials are tied to the user who creates them, ensuring controlled and traceable access.
• Credential Management: Revoke unused credentials at any time from the MCP credentials screen.

Key Benefits

• Connect AI tools and automated workflows to Staircase AI data without browser-based authentication.
• Maintain secure access with short-lived tokens, user-scoped credentials, and easy revocation
• Enable real-time querying of customer data for smarter, data-driven decisions
• Simplify setup for programmatic use cases, reducing friction for teams building AI-powered workflows

For more information, refer to the Create Automations Using the Staircase AI MCP Server article.  

Detect Events with Greater Accuracy Using the Upgraded AI Engine

Staircase AI has rebuilt the underlying language model and prompt architecture that powers Lifecycle Events. The upgraded engine detects events with greater precision, reduces false positives, and improves reliability across all event types. It now sources signals from all communication types for all events.

What’s New

• Control Event Detection with Sub-Signal Tuning: Admins can now view and manage the individual sub-signals that make up each event type, giving them granular control over what triggers an event. Additionally, they can use the checkbox in the Included Signals section of each event type, to turn on or turn off sub-signals as needed.
  
  

• Rename Event Types to Match Your Organization's Terminology: Admins can now rename any out-of-the-box event type to align with the language their team uses from Label field. Event names appear on the account timeline, in reports, and in notifications. Renaming events to familiar terms helps teams recognize and respond to signals faster.
  
  
   
• Review Signal Details and AI Explanations on Timeline:  When users open a Lifecycle Event, the slide-out panel now displays the signal types that fired, an AI-generated explanation (when an explanation is available) of why the event was triggered, and the exact customer quote that caused it. This helps teams validate events and understand root cause without reading the full communication thread.
  
  

• Analyze Signal Details Across Accounts in Lifecycle Events Report: The Lifecycle Events report includes two new columns—Signals and Direct quotes to identify patterns across your portfolio without opening individual accounts.
  
  

For more information, refer to the Lifecycle Events and Lifecycle Events Report articles.