Cognigy

Use when adding custom logic inside a Cognigy tool branch with manage_flow_nodes — supported node types, config schemas, placement rules, and the tool-first workflow.

Use when the user wants to add xApps to a Cognigy flow — interactive HTML / Adaptive Card mini-apps rendered to the user, with the submitted data flowing back into the conversation. Covers the xApp flow nodes, the Init Session rule, and accessing xApp data.

Add a new MCP tool (or extend an existing one) in the NiCE Cognigy Plugin's MCP server

Use when the user wants to build, create, or set up a new Cognigy AI Agent from scratch — covers listing projects, ensuring an LLM exists, creating the agent, testing it, and refining persona/job fields.

Use when the user wants to add knowledge, RAG, or a knowledge store/source to a Cognigy agent — covers embedding vs Knowledge Search models, Knowledge AI settings, ingesting sources, and attaching knowledge as a tool.

Use when configuring or choosing an LLM for a Cognigy agent — valid provider names (openAI, anthropic, azureOpenAI, google, mistral), model strings, connection types, and credential resolution.

Use when exporting, importing, uploading, inspecting, or downloading Cognigy package zip files — including reusing an LLM plus its connection across projects.

Use when configuring project-level Cognigy settings — voice preview / speech provider configuration and Knowledge AI settings.

Validate that a Cognigy Extension's source code follows the repo's required structure BEFORE a pull request is confirmed/merged. Use when the user asks to check/validate/review an extension's structure, prepare an extension for a PR, or confirm an extension is structured correctly. Inspects only raw source — the extension folder layout, the .ts files under src/, and package.json — then reports in plain text what is wrong and shows an example of the correct form so the contributor can fix it. Does NOT inspect build/ output, node_modules, or run the code.

Guide for adding a new Knowledge Connector to a Cognigy Extension. Knowledge Connectors enable Cognigy AI users to fetch data from external knowledge bases (like Confluence, SharePoint, wikis) and create Knowledge chunks for storage in vector databases to be used by AI Agents. This skill covers file structure, implementation patterns, and best practices.

Guide for adding unit tests to a Knowledge Connector in a Cognigy Extension. Covers project setup (dependencies, scripts, build exclusion), test scaffolding using the Node.js standard test library (node:test + node:assert), the standard KnowledgeApi mock pattern, vendor-specific API mocking strategies, required test scenarios, and verification steps.

Cut, build, and ship a new @cognigy/webchat release in this repo — version bump, dependency updates (axios, @cognigy/chat-components, @cognigy/socket-client), OSS license regeneration, the release PR, tagging, the Automatic Draft Release workflow, the changelog, and publishing as a pre-release to npm. Use whenever the user wants to release Webchat, cut/prepare a new version, bump the version, "do a webchat release", update a core dependency and ship it, create or fill a draft release, publish a pre-release, or asks about the release/draft/publish CI workflows. Also use when diagnosing release CI failures (E2E specs broken by a dependency bump, Format/Prettier check, OSS license version mismatch).

Build or modify accessible (WCAG 2.2 Level AA) Webchat UI in this repo. Use whenever creating or editing any React component, control, button, icon button, link, input, textarea, file upload, persistent menu, home screen, header, modal/dialog, overlay, lightbox, tooltip, message renderer, or any JSX/markup — and whenever working on keyboard navigation, focus management, ARIA, screen-reader announcements, live regions, color contrast, or aria-labels. Provides Webchat-specific accessibility recipes and a pre-finish self-check.

Build or modify accessible (WCAG 2.2 Level AA) components in @cognigy/chat-components. Use whenever creating or editing any React component, message renderer, button, link, dialog, datepicker, lightbox, gallery/carousel, audio/video control, adaptive card, or any JSX/markup — and whenever working on keyboard navigation, focus management, ARIA, screen-reader announcements, live regions, or aria-labels. Provides this repo's accessibility recipes, reference implementations, and a pre-finish self-check.

Add a new MCP tool (or extend an existing one) in the cognigy-mcp server