Jul, 134 min to read
Automation
News · Google adds a no-code Agent Designer and Chrome search entry point to Agentspace
Apr, 94 min to read
AutomationGoogle adds a no-code Agent Designer and Chrome search entry point to Agentspace
At Cloud Next, Google expanded Agentspace with employee-built agents, prebuilt Deep Research and Idea Generation agents, and enterprise search inside Chrome — a product that has since been folded into Gemini Enterprise.
What Google actually shipped into Agentspace
The April 9 announcement bundles three concrete changes. First, employees can now reach Agentspace's unified enterprise search, analysis and synthesis capabilities directly from the search box in Chrome. Second, a no-code Agent Designer lets employees of any technical skill level build custom agents for their own workflows. Third, Agentspace ships two Google-built agents, Deep Research and Idea Generation, alongside the already-available NotebookLM for Enterprise.
None of these are model releases. They are distribution and creation surfaces layered on top of Google's foundation models. The stated goal is narrow and operational: make creating and adopting agents simpler. That framing matters because it signals Google is competing on the last mile — how agents get built and reached — rather than on raw model capability alone.
Putting search in the address bar is the automation lever
The Chrome search box integration is the most consequential detail for automation, and it is easy to underrate. Enterprise search tools usually fail not on retrieval quality but on adoption: employees won't leave the tools they already live in to visit a separate portal. By making Agentspace queryable from the browser bar employees already use hundreds of times a day, Google removes the context switch that kills internal knowledge tools.
For teams automating knowledge work, the location of the entry point is as important as the underlying capability. A synthesis engine that requires a deliberate visit gets used weekly; one that answers from the address bar gets used reflexively.
No-code Agent Designer shifts who builds automation
The Agent Designer targets employees of all technical skill levels, meaning the intended author of an agent is the person who owns the workflow, not a central engineering team. That is a deliberate trade. It multiplies the number of agents an organization can produce, but it also distributes responsibility for scoping, data access, and correctness to people who may not think in those terms.
The prebuilt agents — Deep Research and Idea Generation — hedge that risk by offering vetted, Google-authored starting points. In practice, most organizations will see a mix: a handful of official agents plus a long tail of employee-built ones. Governance of that long tail is the work the announcement doesn't describe.
The named customers and the six-month reframing
Google cites interest from Banco BV, Cohesity, Gordon Food Services, KPMG, Rubrik and Wells Fargo — a list weighted toward regulated finance and data-management firms, exactly the environments where enterprise search and controlled agent creation carry the highest stakes and the highest scrutiny.
The most telling line, though, sits in the editor's note dated October 9, 2025: Agentspace is now part of Gemini Enterprise, with its agent creation and orchestration technology powering that platform's core functionality. The features announced here outlived the product name. For anyone evaluating Agentspace on its April terms, the practical implication is that the Chrome entry point, Agent Designer and prebuilt agents were early components of a larger platform consolidation — and should be assessed as Gemini Enterprise capabilities, not a standalone product.
Find this story relevant to you?
Contact us to find a unique solution
Need an AI engineering partner that can actually build?
We help businesses integrate AI, build AI-powered products, automate high-value workflows, and modernize the software systems behind them.
Related reading
More analysis around product delivery, operational AI, and the systems work that makes deployment hold up in reality.