update
May 14, 2026
By Teun
Freshworks adds no-code AI agents to Freshservice
Freshworks has unveiled Freddy AI Agent Studio, a no-code tool for building and deploying AI agents in Freshservice, along with an MCP Gateway and new AI Insights with xLAs. The company says the features aim to move agents into production faster and improve how IT teams measure employee experience.
Freshworks Inc. on Tuesday unveiled a broader set of agentic AI features for its Freshservice IT service management platform, led by a new no-code Freddy AI Agent Studio. The company announced the products at its annual Refresh conference and said the goal is to help enterprises build, customize, and deploy AI agents across service workflows without heavy engineering work.
The new release also includes an MCP Gateway for connecting Freddy AI to third-party tools, plus an analytics layer called AI Insights with Experience Level Agreements, or xLAs. Freshworks said the additions are intended to give customers a single foundation for what it calls “service transformation,” tying together service, asset, and incident data so AI agents can act on it without the cleanup often needed on older platforms.
⚡ New to this?
This matters because many companies want AI agents to do real work, not just answer questions in demos. Freshworks is adding tools to build those agents, connect them to other business software, and measure whether they improve service quality. An SLA, or service-level agreement, is a target for response or resolution time; an xLA adds a view of employee experience and sentiment.
🦞 OpenClaw angle
If you are building self-hosted agents, pay attention to the three parts Freshworks is bundling here: agent creation, tool access, and governance. Treat those as separate layers in your own stack, with clear controls for what each agent can do, which systems it can call, and how changes are approved.
Also, reduce custom glue where you can. Freshworks is betting that standardized connectors and a common context layer speed up deployment, so for your own automations, prefer reusable connectors and a shared event schema over one-off integrations for every workflow.
Finally, add measurement before rollout. If you are deploying agents into IT or employee support flows, track both service metrics and user experience signals so you can tell whether the automation is actually improving outcomes or just shifting work around.
Freshworks is pitching the package as a way to move from pilot projects to production faster. The company said customers should be able to deploy agents in weeks rather than quarters, a claim that reflects the broader pressure enterprise software vendors face as they try to turn AI demos into working systems.
Freddy AI Agent Studio is the centerpiece of the announcement. According to Freshworks, teams can use the no-code environment to create custom agents or start with prebuilt templates, then push those agents into Microsoft Teams, Slack, or employee portals.
The company said the studio also connects with human resources information systems, including platforms from Workday Inc. and Rippling People Center Inc. That lets the agents handle workflows such as onboarding and payroll requests. Freshworks said the studio includes governance controls for managing agents at scale, along with a library of prebuilt agentic workflows that teams can extend.
The MCP Gateway is aimed at a common enterprise problem: connecting AI agents to the rest of the software stack without custom integration work. Freshworks said the gateway allows Freddy AI to pull context directly from tools made by Notion Labs Inc., ClickUp Inc. and Linear Orbit Inc., reducing the need for bespoke code that can slow down rollouts.
Freshworks also introduced AI Insights and xLAs as a measurement layer. Traditional service-level agreements, or SLAs, track metrics such as response and resolution times. Freshworks said xLAs use weighted computation and AI analysis to connect service performance to employee sentiment, giving leaders a broader view of how IT operations affect the employee experience.
The company said the features are supported by telemetry from its customers. Freshworks said analysis of millions of service interactions found that 47% of IT tickets are now submitted outside standard business hours. It also said after-hours response times lag by an hour or more, while SLA compliance can fall by as much as 5%.
Freshworks described that mismatch as a “ghost shift” problem in distributed workforces, where IT teams are effectively serving employees long after normal business hours. The company said its ServiceOps foundation, which includes a redesigned IT asset management product and FireHydrant incident management, provides the base for the new Freddy AI capabilities.
“The true measure of AI’s value isn’t what it can do, it’s what it gives back: time, focus and the freedom for teams to stop fixing yesterday’s problems and start building what’s next,” said Chief Product Officer Srini Raghavan. “Our unified ServiceOps foundation, activated with Freddy AI Agent Studio, is the antidote.”
The announcement comes as vendors including ServiceNow Inc., Atlassian Corp. and Salesforce Inc. continue pushing their own agent platforms. Freshworks is betting that a unified data layer, no-code deployment, and built-in governance will help it stand out as enterprise buyers move more cautiously from AI pilots into production.
