Kore AI Release Notes

<p className="rn-heading">Agent builder experience</p> <p className="rn-subtitle">Control what the model sees from tool results</p>

You can now decide how much of each tool result the model keeps in the conversation: the full result, selected fields only, or a short summary built from a template. Flow steps and completions also accept a note describing what the customer received, for later reasoning. The customer-facing response doesn't change, and large tool payloads no longer crowd out the rest of the conversation history.

<p className="rn-subtitle">Idempotency header for HTTP tools is now opt-in</p>

HTTP tools no longer send an idempotency header on every call by default, which resolves errors from providers such as Salesforce that reject unsupported headers. If your provider supports idempotency, such as Stripe, Adyen, or PayPal, enter its header name in the new Idempotency Header field in the tool's resilience settings. The platform still protects retried calls from running twice.

<p className="rn-subtitle">Filtered bulk delete in Agent Tables</p>

Agent Tables can now delete many rows at once by filtering on any column, not only a unique key. Grant the new bulk delete operation on a table tool, run DELETE ... WHERE in the query console, or use the API. A configurable row limit of up to 10,000 rows guards against accidental large deletes.

<p className="rn-subtitle">Duplicate-run protection for workflow and MCP tools</p> …

<p className="rn-heading">Agent builder experience</p> <p className="rn-subtitle">Agent Tables on non-production deployments</p>

Agent Tables can now be used from deployments in every environment, not only production. All deployment environments currently read and write the same production rows; per-environment table data is not yet available.

<p className="rn-heading">Voice experience</p> <p className="rn-subtitle">Background audio behavior with barge-in disabled</p>

Background audio settings on Pipeline Voice channels now indicate when the audio actually plays. Between-sentence background audio requires barge-in to be enabled on the channel and a streaming text-to-speech voice. When barge-in is disabled, background audio plays only before the agent starts responding, because the reply is spoken as one continuous block with no gaps between sentences.

<p className="rn-subtitle">Caller identity on outbound calls</p>

On outbound calls, the customer is now identified by the number the platform dialed, so contact, memory, and conversation history are attributed to the right person.

<p className="rn-subtitle">Emotech hotwords and Deepgram custom vocabulary controls</p>

Custom vocabulary controls on voice channels now adapt to the speech provider you select:

• Emotech channels add a Hotword Weight field that applies to every hotword in the list, with a recommended range of 5–10. Clearing the list turns hotwords off. …

<p className="rn-heading">Agent builder experience</p> <p className="rn-subtitle">Higher scenario and persona limits for evaluation suites</p>

Evaluation suites can now include up to 500 scenarios and 10 personas, up from 3 and 3, letting you cover far more conversation paths in a single suite. The evaluator limit stays at 12 per suite, and suites generated automatically with Arch still start at 3 scenarios and 3 personas by default.

<p className="rn-subtitle">Unified human task handling</p>

Process workflows can now handle approvals and data collection through a single Human Task node. It pauses the workflow, notifies assignees, and can collect structured data through a reusable Form before resuming based on the selected outcome. You can define SLAs, configure reminders and escalation steps, and specify how to handle SLA breaches. Forms can be created once as reusable, project-level assets and used across workflows. This gives process workflows a simpler way to manage human decisions, data collection, and task deadlines while reducing duplicate form creation.

<p className="rn-subtitle">Scenario-level session variables in Evals</p> …

<Badge> Minor Release </Badge>

This update includes enhancements and bug fixes. The key enhancements included in this release are summarized below.

<p className="rn-heading">UI Enhancements</p><p className="rn-subtitle">Custom user context for Search AI</p>

Admins can now configure which user context fields Agent AI passes to Search AI. The User Context setting in Pass Metadata now lets admins pass Full User Context or Custom Data Only. Learn more→

<p className="rn-subtitle">Agent Coaching drill-down in the Agent AI dashboard</p>

The Agent Coaching section in the Agent AI dashboard includes a rule-level drill-down view. Select a coaching rule from the View All screen to view all instances where the rule triggered, including the trigger text, conversation ID, triggered by, date and time, and GenAI model (for GenAI triggers only). Learn more→

<p className="rn-heading">Integration Enhancements</p><p className="rn-subtitle">NICE Integration with Salesforce Service Cloud</p>

Agent AI integrates with NICE and Salesforce Service Cloud to support transcription, Agent Playbooks, Agent Coaching, Agentic Copilot, automation, and summaries.

<p className="rn-subtitle">Salesforce AppExchange listing correction</p> …

<Badge>Minor Release</Badge>

This update includes enhancements and bug fixes. The key enhancements included in this release are summarized below.

<p className="rn-heading">Console</p><p className="rn-subtitle">Disposition Search in the Console</p>

A search bar in the Add Disposition dialog now lets agents filter dispositions by name as they type, supporting partial and case-insensitive matching. Learn more→

<p className="rn-subtitle">Email Delivery and Attachment Failure Notifications in the Console</p>

Agents now receive a real-time error toaster when an outbound email fails to deliver, or an attachment fails to send, due to SMTP timeouts, server rejections, or network errors. The failed message or attachment is also marked with a failure indicator in the conversation log, replacing the previous ambiguous sent state. Learn more→

<p className="rn-subtitle">Configurable User Information Section in the Console</p>

Administrators can now configure the User Info section in the Console layout, including adding, removing, enabling, disabling, and reordering fields across User Details, Recent Visit History, and Connection Details. Existing layouts retain their current default fields until modified. Learn more→ …

This update includes new features and feature enhancements summarized below.

<p className="rn-title">Guardrails for MCP Server Tools</p>

The Tools section of an MCP server now includes an Enable guardrail toggle that runs account guardrails before any tool executes. Learn more→

<p className="rn-title">Model Management in Workspace Settings</p>

Administrators can now scope the LLM models available to each workspace. Use the new Models tab in Workspace Settings to enable, disable, and reorder models. Learn more→

<p className="rn-title">Desktop Application</p>

AI for Work is now available as a desktop application, giving you access to the assistant, agents, and search without opening a browser.

<p className="rn-title">Live Response Flow</p>

The assistant now streams its thoughts live in a new response flow while it processes your query.

<p className="rn-title">Attachments Grouped with Parent Pages in Enterprise Search</p>

Enterprise Search results now show attachments within their parent page result instead of listing them as separate results.

<p className="rn-title">Search Platform Integration</p>

AI for Work now integrates with the Artemis-based Search Platform, so Enterprise Search and search agents use the updated search infrastructure.

<p className="rn-title">Follow-up Queries for Generated Images</p> …

<p className="rn-heading">Knowledge base</p> <p className="rn-subtitle">PDF formula extraction</p>

Knowledge bases can now extract mathematical formulas from PDF documents. Previously, formulas weren't recognized as either text or images during extraction, so they were left out of the indexed content. They're now captured along with the rest of the document. This is enabled by default for newly created knowledge bases, improving retrieval quality for technical and scientific content. Existing knowledge bases keep their current extraction configuration.

<p className="rn-subtitle">SharePoint connector: quick actions and permission-aware search</p>

The SharePoint connector's overview screen in Studio now includes a Quick Actions bar for Sync, Pause, Resume, Edit Config, Re-Auth, Health Check, and Search, all directly from the connector overview. <br />Learn more →

<p className="rn-heading">Voice experience</p> <p className="rn-subtitle">Gemini Live: configurable context-window compression</p>

For long voice conversations on Gemini Live, you can now configure context-window compression directly from the channel's configuration panel in Studio — turn compression on or off, set the token count at which it triggers, and the token count it compresses down to. This helps longer conversations retain relevant context without hitting the model's context limit.

<p className="rn-heading">Voice experience</p> <p className="rn-subtitle">Contact Centre lifecycle events from CCAI</p>

The platform now derives Contact Centre lifecycle events (agent connected, agent disconnected, ACW completed) from CCAI's conversation_info payload, removing the need for explicitly pushed event entries. Canonical transfer_initiated events are now emitted for voice SIP transfers. The dispositionCode and wrapUpNotes field mapping has been corrected, and close events now include queue and assignment fields.

<p className="rn-subtitle">Language-based voice channel routing</p>

Callers can now select their preferred language at the start of a call (via DTMF, for example, press 1 for English, 2 for Spanish) and are routed to a language-specific voice channel with matching ASR/TTS settings. The selected language is retained for the entire session, preventing the agent from unexpectedly switching between languages mid-conversation.

<p className="rn-subtitle">Runtime-driven dynamic language switching</p> …

<p className="rn-heading">Agent builder experience</p> <p className="rn-subtitle">Comment-only survey templates and simplified follow-up rules</p>

Feedback templates in Studio can now be authored without a numeric rating, so you can collect comments on their own. Per-rating follow-up questions can also inherit their visibility rule from the survey's primary control, instead of being configured separately.

<p className="rn-heading">Voice experience</p> <p className="rn-subtitle">Per-turn TTS delivery policy for protected voice responses</p>

Flow authors can now configure the STREAMING delivery mode for voice responses, alongside the existing barge_in control. This lets authors protect important spoken responses — such as compliance disclosures or handoff completion messages — from being interrupted, truncated, or affected by tool calls, reconnects, or mid-call capability changes. <br />Learn more →

<Badge> Patch Release </Badge>

This update includes enhancements and bug fixes. The key enhancements included in this release are summarized below.

<p className="rn-heading">Widget Enhancements</p><p className="rn-subtitle">Welcome messages for auto-transferred calls</p>

When you set Answer Mode to Auto and a call transfers to the next available agent due to timeout or unavailability, the Agent AI widget shows welcome messages to the newly assigned agent.

<p className="rn-subtitle">Automatic playbook step adherence for IVR transfers</p>

For IVR-to-agent call transfers, Agent AI now automatically marks completed playbook steps as adhered to, based on the information it collects during the IVR-user conversation. This applies to both primary and dynamic playbooks.

<p className="rn-subtitle">Utterance analysis for coaching rules</p>

The Analyse option for coaching rules helps optimize utterance matching through overall or custom analysis. You can identify ambiguous utterance pairs, review threshold summaries, or analyze up to 10 custom utterances and their match scores.

<p className="rn-heading">API</p><p className="rn-subtitle">Agent AI Engagement Tracking and Export APIs</p> …

<Badge> Patch Release </Badge>

This update includes enhancements and bug fixes. The key enhancements included in this release are summarized below.

<p className="rn-heading">Configuration</p><p className="rn-subtitle">End Flow Node Message Optional</p>

The End Flow node message field is now optional, letting flows close without sending a placeholder message to the customer. Learn more→

<p className="rn-heading">Campaigns</p><p className="rn-subtitle">Campaign Attempt Metadata Sent to Artemis for All Call Attempts</p>

For every call attempt - whether or not the customer was reached - Campaigns now sends Artemis the campaign ID, name, instance ID, instance start time, attempt number, and attempt reason. The reason identifies calls placed after a customer requested a callback later, enabling Artemis to distinguish these from standard call attempts. Learn more→

<p className="rn-subtitle">Dynamic Date Constants in Contact List Filters</p>

Contact List filters now support TODAY, TOMORROW, and YESTERDAY as dynamic values for date fields, along with a new DATE_EQUALS operator for date-only equality comparisons. Comparisons use each contact's configured timezone; contacts without a timezone are skipped. Learn more→

---

<p className="rn-heading">Voice experience</p> <p className="rn-subtitle">Amazon Nova Sonic speech-to-speech voice</p>

Voice channels can now use Amazon Nova Sonic, a speech-to-speech (S2S) model served through AWS Bedrock, as a realtime voice provider. Configure it in Studio's Voice Services page, using either access-key or IAM-role-based authentication.

<p className="rn-subtitle">Vertex AI Gemini Live for realtime voice</p>

Realtime LLM voice channels can now be powered by Vertex AI's Gemini Live models, in addition to the direct Google Gemini Live provider. Model resolution and credential handling for Vertex use the same Voice Services configuration flow as other realtime providers.

<p className="rn-subtitle">Native multilingual voice profiles</p>

Voice channels now support per-language voice profiles: each supported interaction language can have its own ASR and TTS provider and voice selection, with the voice runtime handling automatic cutover and fallback. Learn more →

<p className="rn-subtitle">Voice CSAT survey on the return leg</p>

Voice transfers over SIP now support a post-agent CSAT survey played back on the call's return leg after a human agent completes the interaction, using a reserved x-callee-number SIP header to identify the transfer target.

<p className="rn-subtitle">GCP authentication for Vertex AI and Gemini</p>

Auth profiles for Vertex AI and Gemini now support: …

<p className="rn-heading">Agent builder experience</p> <p className="rn-subtitle">Per-server rate limits for MCP tools</p>

Rate limits can now be configured per MCP server, with Studio controls to set the limit and automatic enforcement across tenants, so one server's traffic can't overwhelm the rest. Learn more →

<p className="rn-subtitle">Automatic indentation validation and repair in the ABL editor</p>

The ABL editor now detects indentation problems as you write and offers an auto-fix, with clearer diagnostics when something can't be repaired automatically.

<p className="rn-subtitle">Link eval suites to a version, deployment, or environment</p>

Eval suites can now run against a specific project version, deployment, or environment. This helps teams determine whether a score change came from a project edit or another change, and evaluate what is actually deployed. <br />Learn more →

<p className="rn-subtitle">Workflow versions, settings, and a redesigned monitor experience</p>

Workflows now support versioning and a dedicated settings tab, alongside a redesigned monitor experience in Studio with clearer run detail, timeline, status, outcomes, and payload views. <br />Learn more →

<p className="rn-subtitle">Move workflows between projects</p> …

<Badge> Minor Release </Badge>

This update includes enhancements and bug fixes. The key enhancements included in this release are summarized below.

<p className="rn-heading">Widget Enhancements</p><p className="rn-subtitle">Manual NBA triggering with proactive mode control</p>

Agents can now select Generate NBA on the Assist tab to manually trigger an NBA suggestion when all of the following apply:

• Proactive Mode is off
• A new customer message arrives
• The previous NBA generation finishes

Learn more→

<p className="rn-subtitle">Call transfer reason for bot-to-human agent transfers</p>

For bot-to-human call transfers, the Landing Summary now includes a Call Transfer Reason section. Learn more→

<p className="rn-subtitle">Transcript visibility control for agents</p>

Admins can now control transcript visibility for transferred agents on the voice channel, using the Control Transcript Visibility between Agents setting in Widget Settings. When admins select Use Agents' Entry and Exit Events, the current agent sees transcripts of their own conversation and summaries of previous bot-user and agent-user conversations. Learn more→

<p className="rn-heading">UI Enhancements</p><p className="rn-subtitle">Coaching nudge enhancements</p> …

<Badge>Minor Release</Badge>

This update includes enhancements and bug fixes. The key enhancements included in this release are summarized below.

<p className="rn-heading">Console</p><p className="rn-subtitle">Revamped Conversation History With Omnichannel Summary</p>

The Conversation History panel in the Agent Console now presents Chat, Email, and Voice interactions in a unified, chronological view. Each conversation displays acceptance time, duration, channel, participating agents, bot involvement, and status codes. Learn more→

<p className="rn-subtitle">Conversation Position Preserved When Switching Between Concurrent Chats</p>

Conversations now open directly - at the last-read message when unread messages exist, or at the latest message when fully read - with no visible scrolling.

<p className="rn-subtitle">Network Instability Warning for One-way Audio</p>

A warning now appears in the Console when audio packets sent or received drop below the threshold indicating one-way audio, alerting agents to check their network or reconnect the call. The warning dismisses when two-way audio is restored, or can be manually closed by the agent.

<p className="rn-heading">Configuration</p><p className="rn-subtitle">Separate Permissions for Configuration and Analytics Wallboards</p> …

<p className="rn-heading">Agent builder experience</p> <p className="rn-subtitle">Support for Base64-encoded file attachments</p>

Alongside the existing multipart/form-data upload path, the platform now accepts Base64-encoded attachments in application/json requests, using filename, mimeType, and contentBase64. Successful uploads can return a short-lived signed delivery URL. If the delivery URL expires, you can request a new one without re-uploading the file.

<p className="rn-subtitle">Workflow Agent node accepts attachment IDs</p>

The Agent node's canvas attachments field now accepts attachment IDs directly, in addition to https URLs, so a workflow can hand an already-uploaded attachment to an agent node without re-hosting it as a URL first. <br />Learn more →