Cognigy/cognigy-plugin

README preview

# NiCE Cognigy Plugin

> Distributed exclusively through each client's native plugin mechanism — a **plugin** on **Claude Code**, **Claude Desktop**, **ChatGPT + Codex**, and **Antigravity**, an **extension** on **Google Gemini CLI**, a **plugin** on **Cursor**, and an [**Agent Plugins**](https://agent-plugins.org)-standard plugin on any conformant host (**Kiro**, **VS Code + Copilot**, …) — with more clients to come. Each package installs the server engine and ships skills + agents.

A plugin that connects your AI assistant to the [Cognigy.AI](https://www.cognigy.com) REST API. Create, test, and improve LLM-based AI Agents through a self-improvement loop — without leaving your client.

**Quick links:** [Installation](#installation) · [Staying up to date](#staying-up-to-date)

## Features

- **17 workflow tools** for agent creation, deployment, packaging, backup, and voice setup
- **One-call agent setup**: creates Agent + Flow + AI Agent Job Node + REST Endpoint automatically
- **Self-improvement loop**: talk to your agent, evaluate responses, update the job description, repeat
- **Knowledge store support**: attach RAG knowledge stores to agents as tools
- **Browser voice deployment**: create Voice Gateway endpoints with WebRTC demo URLs
- **Voice preview setup**: configure supported speech providers for voice experiences
- **Skills + agents**: workflow guidance auto-loads as skills in supporting clients; build/go-live loops run as subagents
- **Built-in docs lookup**: bundles the official [Cognigy documentation](https://docs.cognigy.com) MCP server — your assistant searches and reads the docs before answering platform questions, instead of guessing from training data
- Built-in rate limiting, Zod input validation, and RFC 7807 error responses

View full repository on GitHub →