# NiCE launches next-generation Cognigy with redesigned studio, skills and tracing for running AI agents

published: 2026-11-14T18:05:27.077Z
topic: infrastructure
source: https://siliconangle.com/2026/11/13/nices-next-generation-cognigy-bets-the-hard-part-of-ai-agents-is-running-them-not-building-them/
url: https://aiagentnews.io/stories/nice-launches-next-generation-cognigy-with-redesigned-studio-skills-and-tracing

NiCE Ltd. has launched the next generation of its Cognigy platform, adding a redesigned Agentic Experience Studio, reusable skills, turn-by-turn tracing and native voice-to-voice model support. The company is positioning the product around operating agents reliably in contact centers rather than around building smarter models.

The release includes an AI Assist companion for building, testing and improving agents, along with reusable skills that package instructions and tools. Default speech-to-text, text-to-speech and large language models are now included in the per-conversation price, while premium models from Deepgram, ElevenLabs and Anthropic can be added with a few clicks and are billed by usage.

NiCE says the new studio lets builders create an agent quickly with models already provisioned, while advanced sections keep full access to the system prompt. The Simulator now supports multivariate testing of up to 10 variants, using an LLM-as-judge to run synthetic customers against an agent. Cognigy also works as both a Model Context Protocol client and server and supports agent-to-agent connections.

NiCE is co-developing the Personal Agent Protocol with Meta and Sierra, which is not yet live. NiCE executives said the release is an evolution rather than a replacement, with existing production deployments and integrations continuing to run unchanged.

## Why it matters if you build agents

Developers building contact center agents can use per-step tracing to see which skill loaded, which tool was called and when a conversation handed off to a deterministic flow.

Teams can compare model variants on cost per resolved conversation using built-in simulation instead of anecdotal results.

Lowering the setup barrier means more agents may be built by less experienced staff, so guardrails and disclosure rules need to be set up deliberately.
