diff --git a/agent-platform/analytics-insights/agent-transfer.mdx b/agent-platform/analytics-insights/agent-transfer.mdx index 40574b3b..1c1f1e4e 100644 --- a/agent-platform/analytics-insights/agent-transfer.mdx +++ b/agent-platform/analytics-insights/agent-transfer.mdx @@ -3,7 +3,7 @@ title: Agent Transfer sidebarTitle: Agent Transfer --- -The **Agent Transfer** page — titled **Queues and Agents** in the interface — provides efficiency and performance metrics for conversations that AI agents escalated to human operators. It helps operations teams monitor transfer volumes, queue wait times, and human-agent performance to ensure that escalated conversations receive timely, high-quality support. +This document covers human escalation metrics in Analytics and Insights: transfer efficiency by channel, queue wait times and abandonment rates, and human-agent performance for escalated conversations. **Navigation**: **Project** → **Insights** → **Agent Transfer** diff --git a/agent-platform/analytics-insights/analytics-pipelines.mdx b/agent-platform/analytics-insights/analytics-pipelines.mdx index ebae5cb8..178de8d4 100644 --- a/agent-platform/analytics-insights/analytics-pipelines.mdx +++ b/agent-platform/analytics-insights/analytics-pipelines.mdx @@ -3,7 +3,7 @@ title: Analytics Pipelines sidebarTitle: Analytics Pipelines --- -The **Pipelines** page is where Insights shifts from pre-built dashboards to user-defined analytics. The platform ships with built-in pipelines covering common evaluation needs — from sentiment analysis to anomaly detection — and provides a visual node-based editor for creating custom pipelines that encode your organization's specific quality criteria. Pipelines are the engine behind Agent Performance, Quality Monitor, and Customer Insights: without active pipelines, those pages have no data to display. +This document covers the pipeline framework in Analytics and Insights: the built-in pipelines that power Agent Performance, Quality Monitor, and Customer Insights, and the visual node-based editor for building custom evaluation logic. **Navigation**: **Project** → **Insights** → **Pipelines** diff --git a/agent-platform/analytics-insights/customer-insight.mdx b/agent-platform/analytics-insights/customer-insight.mdx index 99f16f4f..18be527e 100644 --- a/agent-platform/analytics-insights/customer-insight.mdx +++ b/agent-platform/analytics-insights/customer-insight.mdx @@ -3,7 +3,7 @@ title: Customer Insights sidebarTitle: Customer Insights --- -This page brings together the analytical view of customer behavior — what they are asking about, how they feel, and whether their issues get resolved — with the raw qualitative feedback they leave after conversations. Use Customer Insights to identify emerging intent patterns and sentiment trends, and Feedback to read what users actually said in their own words. +This document covers the customer experience layer in Analytics and Insights: the Customer Insights page for intent classification, sentiment trajectory, and frustration detection, and the Feedback page for end-user ratings and verbatim comments. ## Customer Insights diff --git a/agent-platform/analytics-insights/executive-dashboard-usage.mdx b/agent-platform/analytics-insights/executive-dashboard-usage.mdx index 1308804c..8d8854be 100644 --- a/agent-platform/analytics-insights/executive-dashboard-usage.mdx +++ b/agent-platform/analytics-insights/executive-dashboard-usage.mdx @@ -3,7 +3,7 @@ title: Executive Dashboard and Usage sidebarTitle: Dashboard and Usage --- -This page covers the core monitoring stack: the executive dashboard for high-level KPIs and trends, the Analytics page for operational drill-downs and session-level debugging, and Billing and Usage for cost governance. Together, these three areas give you a complete picture of what is happening across your AI agent program right now. +This document covers the core monitoring stack in Analytics and Insights: the Executive Dashboard for high-level KPIs and trend analysis, the Analytics page for operational drill-downs and session-level debugging, and Billing and Usage for cost governance. ## Dashboard diff --git a/agent-platform/analytics-insights/performance-quality.mdx b/agent-platform/analytics-insights/performance-quality.mdx index 0dc7283c..9d467d72 100644 --- a/agent-platform/analytics-insights/performance-quality.mdx +++ b/agent-platform/analytics-insights/performance-quality.mdx @@ -3,7 +3,7 @@ title: Performance and Quality sidebarTitle: Performance and Quality --- -This page covers two closely related evaluation surfaces: Agent Performance, which scores individual agents across quality dimensions, and Quality Monitor, which tracks system-wide quality health. Both are powered by the same analytics pipelines and evaluate the same five dimensions — quality, faithfulness, knowledge coverage, safety, and context preservation. The typical workflow is to spot a problem in Quality Monitor and then switch to Agent Performance to identify which agent is responsible. +This document covers agent evaluation in Analytics and Insights: the Agent Performance page for per-agent quality scorecards and comparison, and the Quality Monitor page for system-wide quality health across five evaluation dimensions. ## Agent Performance diff --git a/agent-platform/analytics-insights/voice-analytics.mdx b/agent-platform/analytics-insights/voice-analytics.mdx index bba90a1a..e90413eb 100644 --- a/agent-platform/analytics-insights/voice-analytics.mdx +++ b/agent-platform/analytics-insights/voice-analytics.mdx @@ -3,7 +3,7 @@ title: Voice Analytics sidebarTitle: Voice Analytics --- -The **Voice Analytics** page provides a dedicated dashboard for monitoring call quality, speech recognition accuracy, and end-to-end latency across the voice processing pipeline. For voice-enabled agents, use this page to ensure audio quality, recognition accuracy, and response latency meet caller expectations, and to diagnose degradations before they affect customer satisfaction at scale. +This document covers voice channel monitoring in Analytics and Insights: call quality scoring, speech recognition accuracy, end-to-end latency tracking, barge-in rates, and DTMF fallback metrics for voice-enabled agents. **Navigation**: **Project** → **Insights** → **Voice Analytics** diff --git a/agent-platform/external-agent.mdx b/agent-platform/external-agent.mdx index 9096085a..911082cb 100644 --- a/agent-platform/external-agent.mdx +++ b/agent-platform/external-agent.mdx @@ -3,11 +3,7 @@ title: "External Agent" sidebarTitle: "External Agent" --- -The Agent Platform supports cross-service agent invocation via the A2A (Agent-to-Agent) protocol. This allows agents deployed on different runtime instances, including agents owned by external organizations, to communicate over standard HTTP. This helps connect agents to other agents, enabling delegation, sub-task hand-off, and multi-agent orchestration across service boundaries. - -The routing executor transparently handles remote handoffs through the A2A client within Agent Platform, making agent topology location-agnostic. A specialized agent running on external infrastructure is called exactly the same as a local agent. - -The platform handles communication with external agents through the A2A protocol, allowing your agents to collaborate seamlessly regardless of where they are hosted. +Register and use external agents in Agent Platform using the A2A (Agent-to-Agent) protocol. This enables cross-service agent invocation over standard HTTP — including agents owned by external organizations — supporting delegation, sub-task handoff, and multi-agent orchestration across service boundaries. The routing executor handles remote handoffs transparently, making agent calls location-agnostic: external agents are invoked the same way as local ones. ## Register an External Agent @@ -15,39 +11,35 @@ Register an external agent to make it available within the Agent Platform. **Prerequisites** -Ensure that: - - The external agent supports the A2A protocol. - The agent endpoint is reachable over HTTP or HTTPS. - You have the endpoint URL for the external agent. - Authentication details are available if required. -**To Register an External Agent** +**To register an external agent:** 1. Navigate to **External Agents**. 2. Click **Register Agent**. 3. Enter the required details: - - Agent Name - Name of the external agent. - - Endpoint URL -URL of the external agent endpoint. - - Protocol - communication protocol for interaction with the external agent. Agent Platform supports: - - A2A - - REST - - Authentication Type - Authentication required for communication. Select the appropriate authentication mechanism based on the requirements of the external agent endpoint. Agent platform supports: + - **Agent Name** — Name of the external agent. + - **Endpoint URL** — URL of the external agent endpoint. + - **Protocol** — Communication protocol for the external agent. Supported protocols: + - A2A + - REST + - **Authentication Type** — Authentication mechanism required by the external agent endpoint. Supported types: - API Key - Bearer Token 4. Click **Register**. -The platform validates the endpoint and adds the agent to the External Agents list. Use the **Test Connection** action to verify that the external agent endpoint is reachable. If successful, it fetches and stores the agent card that defines the capabilities, skills, streaming support, etc. for the external agent. +The platform validates the endpoint and adds the agent to the **External Agents** list. Use the **Test Connection** action to verify that the endpoint is reachable. On success, the platform fetches and stores the agent card, which defines the agent's capabilities, skills, and streaming support. -## Access External Agents via Handoff or Delegation +## Access External Agents via Handoff or Delegation -Once registered, external agents become available as tools that can be used within Agent Platform agents. You can combine local and external agents in the same multi-agent workflow. +Once registered, external agents are available as tools within Agent Platform agents. You can combine local and external agents in the same multi-agent workflow. -To add an external agent, add it to the Handoff or Delegate section of the agent. The platform automatically handles communication with the external agent through the A2A protocol. +To use an external agent, add it to the **Handoff** or **Delegate** section of your agent. The platform handles communication through the A2A protocol. -**Via ABL**: - -Add the external agent under the HandOff or Delegate sections and specify the location as `remote`. +**In ABL**, add the external agent under the `HANDOFF` or `DELEGATE` section and set `LOCATION` to `REMOTE`: ```yaml HANDOFF: @@ -74,14 +66,14 @@ HANDOFF: | `PROTOCOL` | `A2A` or `REST` | The communication protocol used to invoke the remote agent. | | `ASYNC` | `true` / `false` | Enables asynchronous dispatch with push-notification barriers. When set to `true`, configure the `CALLBACK_BASE_URL` environment variable. This value must be a publicly accessible HTTPS URL where the remote agent can post execution results. | | `TIMEOUT` | Integer (seconds) | Maximum time to wait for a response from the remote agent before the request times out. | -|`CONTEXT`|-|Copies named fields from the current session's gathered/set data values into the request metadata sent to the external agent| -|`RETURN`|`true/false`|Whether the parent thread waits for response or not. | +| `CONTEXT` | — | Copies named fields from the current session's gathered or set data values into the request metadata sent to the external agent. | +| `RETURN` | `true` / `false` | Whether the parent thread waits for the response. | ### Operating Modes #### Sync Mode -This is the default mode. Agent waits for a response from external agent. Default timeout value is 30 seconds. If the request times out, `ON_FAILURE` event is triggered. +This is the default mode. The agent waits for a response from the external agent. The default timeout is 30 seconds. If the request times out, the `ON_FAILURE` event is triggered. ```yaml - TO: Risk_Analyzer @@ -97,7 +89,7 @@ This is the default mode. Agent waits for a response from external agent. Defaul #### Streaming -This is auto-detected when remote agent card publishes `streaming: true` and user is connected. No change is required in DSL. The runtime handles it automatically. +Streaming is auto-detected when the remote agent card has `streaming: true` and the user is connected. No DSL changes are required; the runtime handles it automatically. ```yaml - TO: Risk_Analyzer @@ -108,15 +100,15 @@ This is auto-detected when remote agent card publishes `streaming: true` and use RETURN: true ``` -#### Async +#### Async -This is based on suspend and callback mechanism. The current session suspends. Remote agent pushes result when done. Used for long-running operations. +Async mode uses a suspend-and-callback mechanism. The current session suspends, and the remote agent pushes the result when done. Use this mode for long-running operations. Requirements for async remote calls: -- CALLBACK_BASE_URL environment variable must be set to a public HTTPS URL that the remote agent can reach. +- The `CALLBACK_BASE_URL` environment variable must be set to a public HTTPS URL reachable by the remote agent. - The remote agent's card must have push notification capabilities. -- Default async timeout is 300 seconds. It can be overridden by the timeout value in the ABL. +- Default async timeout is 300 seconds. Override with the `TIMEOUT` value in ABL. ```yaml @@ -133,10 +125,9 @@ Requirements for async remote calls: RETURN: true ``` -### Points to Note: - -1. **Case sensitivity**: `TO: ` must exactly match the name field provided during registration. It is case-sensitive. -2. **Config variable**: If the URL is provided using config fields, the variable` {{config.EXTERNAL_AGENT_URL}}` is resolved at compile time, not runtime. The config variable must exist in the project before you save and deploy the agent. This is recommended for multi-environment set ups. + +1. **Case sensitivity**: `TO: ` must exactly match the name provided during registration. The value is case-sensitive. +2. **Config variable**: When using a config variable for the URL, `{{config.EXTERNAL_AGENT_URL}}` resolves at compile time, not runtime. The config variable must exist in the project before you save and deploy the agent. This approach is recommended for multi-environment setups. ```yaml - TO: Hosted_External_Agent @@ -148,7 +139,8 @@ Requirements for async remote calls: history: full RETURN: true ``` -3. **ASYNC: true**: Push-notification callbacks need a public `HTTPS CALLBACK_BASE_URL` env variable. -4. Inline `ENDPOINT` in DSL overrides the endpoint provided during registration, but auth details must be provided during registration. You can't put credentials in the DSL. -5. Remote agents are stateless by default - they only see the current message. Use `history: full` in `CONTEXT:` to forward the conversation. The history is stored in `message.metadata.history` in the A2A payload. -6. A `contextId` is sent on every turn to the same remote agent, allowing the remote side to associate turns if it keeps state. +3. **Async callback**: Push-notification callbacks require a public HTTPS URL set in the `CALLBACK_BASE_URL` environment variable. +4. **Inline endpoint**: An inline `ENDPOINT` in DSL overrides the endpoint provided during registration. Authentication details must still be configured during registration—credentials cannot be added to the DSL. +5. **Remote agent state**: Remote agents are stateless by default and only see the current message. Use `history: full` in `CONTEXT` to forward the conversation. The history is stored in `message.metadata.history` in the A2A payload. +6. **Context tracking**: A `contextId` is sent on every turn to the same remote agent, allowing the remote side to associate turns if it maintains state. + diff --git a/agent-platform/index.mdx b/agent-platform/index.mdx index b987bce0..56b2c039 100644 --- a/agent-platform/index.mdx +++ b/agent-platform/index.mdx @@ -1,5 +1,5 @@ --- -title: 'Agent Platform Overview' +title: "Agent Platform { Artemis } Overview" sidebarTitle: 'Platform Overview' --- An AI-native platform for building, running, and governing programmable agent systems at enterprise scale. diff --git a/agent-platform/release-notes/recent-updates.mdx b/agent-platform/release-notes/recent-updates.mdx index 4e9b2ebf..dae54bf1 100644 --- a/agent-platform/release-notes/recent-updates.mdx +++ b/agent-platform/release-notes/recent-updates.mdx @@ -1,9 +1,10 @@ --- -title: Agent Platform Release Notes -sidebarTitle: Release Notes +title: "Agent Platform { Artemis } Release Notes" +sidebarTitle: 'Release Notes' --- -This document provides information on the feature updates and enhancements introduced in the recent Agent Platform \{ Artemis \} releases. +This document provides information on the feature updates and enhancements introduced in the recent Agent Platform **\{ Artemis \}** releases. + ## v1.1.1 June 05, 2026 @@ -23,6 +24,14 @@ Simplified the SearchAI Knowledge tool DSL, improved knowledge-tool binding and Added support for the `effect` and `provides` properties in AgentTool definitions and the `satisfiedBy` property in GatherField definitions. +--- + +## v1.1.0 June 01, 2026 + +Minor Release + +This update includes new features and enhancements summarized below. +

Credential Management Updates

Introduced Config Variables (`{{config.X}}`) and Auth Profiles as the new credential management model, replacing Tool Secrets and the Tool Secrets Management API. @@ -128,8 +137,6 @@ Fixed a regression on the agent canvas where delegate relationship edges were dr The Behavior Profile CONVERSATION block now applies all configured fields, resolving a case where only style and tone were applied and other fields were silently dropped. ---- -

Tools and Workflows

Tools page "New Tool" menu fix

@@ -140,24 +147,18 @@ The Studio Tools page "+ New Tool" continues to be visible when the current tab Fixed an issue where ABL workflows didn't trigger when invoked from the AI for Work (AI4W) channel. ---- -

Insights and Dashboards

Agent Transfer dashboard

Added an Agent Transfer insights dashboard. You can now analyze how and when conversations are transferred between agents. ---- -

Voice

Reliable voice channel provisioning

Fixed an issue to improve the reliability of voice channel setup, during Jambonz WSS application provisioning. ---- -

Studio and Administration

Consistent Audit Log date formatting

@@ -168,8 +169,6 @@ Audit Logs now use a consistent date format between the From–To filter and the Git-sync failures are now surfaced clearly in Studio, and import-error messages are normalized for easier troubleshooting. ---- -

Security and Privacy

SDK origin allowlist enforcement

@@ -235,8 +234,6 @@ Guardrails configured on an agent are now reliably applied at run time, closing [Learn more →](/agent-platform/safety-and-guardrails) ---- -

Knowledge and Search AI

Clearer web-crawl ingestion

@@ -263,8 +260,6 @@ Fixes to web-crawl ingestion (including docling filename/extension handling) imp [Learn more →](/ai-for-service/searchai/content-sources) ---- -

Voice

Unified voice provider registry and capability matrix

@@ -279,8 +274,6 @@ Gemini Live voice-activity-detection (VAD) sensitivity is now exposed in the spe [Learn more →](/agent-platform/administration/ai-configuration#voice-services) ---- -

Channels, Contact Center, and Connectors

Contact-Center "After Call Work" (ACW) handling

@@ -301,8 +294,6 @@ A new connector category groups Zendesk and ServiceNow under Service Management. [Learn more →](/ai-for-service/searchai/connectors/connector-directory) ---- -

LLM and Models

AWS Bedrock provider integration

@@ -311,8 +302,6 @@ Adds AWS Bedrock as an LLM provider with IAM role support and additional connect Learn more about [AI Configuration →](/agent-platform/administration/ai-configuration) · [Supported AI Models →](/agent-platform/supported-models) ---- -

Studio, Insights, and Analytics

Insights Auditing — automated metric analysis & recommendations

@@ -331,8 +320,6 @@ A recent-feedback table with filters in Studio to review captured user feedback. Refreshed Studio navigation and a standardized design system, including sidebar localization labels. ---- -

Security, Administration, and Runtime

Platform domain allowlist and DB-managed Platform admins

@@ -353,8 +340,6 @@ Responses now carry metadata indicating whether a reply was LLM-generated or scr [Learn more →](/agent-platform/api-reference/conversation-api) ---- -

Tools and WebSDK

URL-encoded request body support

@@ -377,6 +362,19 @@ The WebSDK now renders hyperlinks, paragraph formatting, and lists instead of sh [Learn more →](/agent-platform/api-reference/sdks) +--- + ## v1.0 May 21, 2026 -Introducing the initial release of Agent Platform \{ Artemis \}, featuring Agent Blueprint Language (ABL™) for agent creation and management, Arch™ AI for agent generation from SOPs, SearchAI for knowledge management and retrieval, web and voice channels, tool integrations, and insights dashboards. For details about the feature, see [Platform Overview](/agent-platform/index). +Kore.ai Agent Platform \{ Artemis \} is now available, with the following capabilities: + +- Agent Blueprint Language (ABL™) — create and manage agents. +- Arch™ AI — generate agents from SOPs. +- SearchAI — manage and retrieve knowledge. +- Channels — deploy agents on web and voice. +- Tool integrations — connect agents to external tools. +- Insights dashboards — monitor agent performance. + +For details, see [Platform Overview](/agent-platform/index). + +--- \ No newline at end of file diff --git a/agent-platform/transfer-sessions.mdx b/agent-platform/transfer-sessions.mdx index ddf5f40a..07de3f5f 100644 --- a/agent-platform/transfer-sessions.mdx +++ b/agent-platform/transfer-sessions.mdx @@ -3,7 +3,7 @@ title: "Transfer Sessions" sidebarTitle: "Transfer Sessions" --- -Transfer Sessions is a monitoring and control page for live handoffs from a bot to a human agent on an external contact center platform. +Transfer Sessions is a monitoring and control page for live handoffs from an AI agent to a human agent on an external contact center platform. Navigate to **Operate** > **Transfer Sessions** from the project sidebar. diff --git a/home.mdx b/home.mdx index 52dbfc0a..26f623d0 100644 --- a/home.mdx +++ b/home.mdx @@ -245,16 +245,16 @@ export const DocsFooter = () => (
-
Kore.ai Agent Platform
+
Kore.ai Agent Platform \{ Artemis \}
-

The AI-programmable foundation for the agentic enterprise, enabling teams to build and deploy AI agents for customer and employee experiences with certainty.

+

An AI-programmable foundation for building and deploying enterprise AI agents across customer and employee experiences, with certainty:

} title="Dual-Brain Architecture" description="A split-brain runtime: an agentic engine that reasons and a deterministic validator, running in parallel over a shared spine." /> - } title="ABL — Agent Blueprint Language" description="One DSL for agents, flows, tools, and policies. Compile-time validated, version-controlled, AI-programmable." /> - } title="Arch AI — The AI That Builds AI" description="A multi-agent system that drives the full lifecycle: interviews stakeholders, designs topology, writes ABL, learns from production." /> + } title="ABL™ — Agent Blueprint Language" description="One DSL for agents, flows, tools, and policies. Compile-time validated, version-controlled, AI-programmable." /> + } title="Arch™ AI — The AI That Builds AI" description="A multi-agent system that drives the full lifecycle: interviews stakeholders, designs topology, writes ABL, learns from production." /> } title="Multi-Agent Orchestration" description="Compose specialists with HANDOFF, DELEGATE, FAN_OUT, and A2A — typed payloads, governed handoffs, unified traces." /> } title="Tools & Integrations" description="Five tool types (HTTP, Code, KB, Workflow, MCP) under one Auth Profile — agents reach every enterprise system without glue code." /> } title="Search & Knowledge AI" description="RAG, vector search, knowledge graph, and reranking — with intelligent pipelines that chunk, embed, and enrich automatically." />