Last updated: 2026-05-13

Source: https://support.freshservice.com/support/solutions/articles/50000012643-introduction-to-ai-agent-studio-in-freshservice

Freddy AI Agent Studio is a platform within Freshservice that lets you build, deploy, and manage AI agents that answer employee questions, create incidents, and take action to resolve service requests. Instead of one generic assistant, you can deploy a workforce of specialized agents for IT, HR, and other business functions. Agents understand requests in context, access the right data, and take action across workflows and systems.

The following video explains what the platform unlocks, how it works, and the key behaviors to watch out for so you can roll it out with confidence:

Build and deploy AI agents without code

What Freddy AI Agent Studio can be used for

You'll get the most value from this platform when you have a steady volume of repeatable employee conversations that today rely on human agents, forms, or one-size-fits-all chatbots. Common reasons to turn it on:

  • Resolve common IT requests autonomously: The ready-made IT Agent ships with pre-built workflows for password resets, software access approvals, and similar high-volume requests.
  • Answer HR policy and process questions: The ready-made HR Agent handles PTO updates, onboarding workflows, and policy queries using your approved content.
  • Build agents for any business function: Use the no-code builder to design custom agents and workflows beyond IT and HR.
  • Deploy on the channels employees already use: Run the same agent across the Support Portal, Microsoft Teams, and Slack, with multi-language support across 40+ languages.

    • Plan availability

    Freddy AI Agent Studio is rolling out for Growth and Pro plans as a promotional offer until September 30, 2026. For more information, see Freddy AI Agent Studio FAQ.

    Key concepts to know before you start

    A few terms appear throughout this article and the rest of the Freddy AI Agent Studio documentation. Knowing them upfront makes setup much easier.

    TermWhat it means
    AgentAn AI agent built in Freddy AI Agent Studio that answers employee questions, creates incidents, and takes action to resolve service requests on one or more channels.
    Pre-built agentsA ready-made agent that ships with the platform. Two are available at launch: the IT Agent and the HR Agent, each with pre-built agentic workflows.
    Agentic workflowA defined sequence of steps that an agent runs to complete an action. For example, resetting a password, approving software access, or updating PTO. 20 pre-built workflows are included at launch.
    Knowledge sourceContent the agent uses to ground its responses: files, URLs, the knowledge base, the service catalog, and external connectors such as SharePoint, Google Drive, and Confluence.
    ChannelWhere the agent is deployed: the Support Portal (web widget), MS Teams, or Slack.
    SessionAny 24-hour period of conversation between a unique user and the agent. A session covers all messages exchanged in that window. If a conversation continues beyond 24 hours, it counts as multiple sessions.

    How agents, knowledge, and workflows connect

    Agents, knowledge sources, and workflows are linked through configuration. Two simple ideas govern the model. Once you know them, the rest of the behavior follows naturally.

  • Agents reason over knowledge: Each agent is grounded in your Freshservice data and any additional knowledge sources you connect, files, URLs, the knowledge base, the service catalog, and apps such as Google Drive, SharePoint, and Confluence. When a user asks a question, the agent retrieves relevant content and composes a grounded answer.
  • Agents take action through workflows: When a conversation needs to do something, reset a password, approve access, or raise a service request, the agent calls a workflow. Workflows can execute or take actions on other systems. 30 app integrations are available currently.

    • Tips to get the most value

  • Start small: Roll out one agent on one channel first, learn from the experience, then expand. It's much easier than re-architecting later.
  • Treat knowledge as the foundation: Your agent is only as good as what it knows. Curate the files, URLs, knowledge base articles, and service catalog content that the agent reads and update them when users raise gaps.
  • Use the out-of-the-box agents as a head start: The IT Agent and HR Agent ship with 20 pre-built workflows. Customizing one of these is usually faster than building a new agent from scratch.
  • Validate with bulk testing before going live: Create test scenarios with sample user queries, run the agent against all scenarios at once, and review results to identify gaps before any real user sees the agent.

    • ``` For answers to common questions about capabilities, pricing, availability, and deployment, see the Freddy AI Agent Studio FAQ. ```