Skip to content
Agent Studio / Planned

On the roadmap · not available to use yet

Give an AI agent a job, a source, and a handoff path.

Agent Studio is planned as a separate place to set up specialist AI agents for customer conversations. The goal is to let a business define what an agent knows, test its answers, and decide when a person should take over.

Planned workflow illustration. This is not a working agent or a live product screenshot.
01 / Why it helps

A clearer way to prepare AI for real conversations.

01

Ground answers in your business

Give a specialist agent relevant policies, product information, and other approved knowledge sources, then show which source supports an answer.

02

Test before release

Try realistic customer questions, inspect the proposed response and action trace, and adjust the setup before an agent handles live chats.

03

Make the human handoff useful

Route an issue to a teammate when the agent cannot answer or a defined situation needs a person, with a concise summary and conversation context.

02 / Planned workflow

From setup to a supervised conversation.

  1. 01

    Define a specialist

    Choose the agent's purpose, connect the knowledge it may use, and set rules for the questions it should handle.

  2. 02

    Test and review

    Check sample conversations, source references, and action traces before deciding whether the agent is ready for deployment.

  3. 03

    Route and monitor

    Send relevant conversations to the agent, keep its work visible in the inbox, and hand exceptions to a person with context.

03 / Availability

Agent Studio is planned

The separate Agent Studio, knowledge sources, testing, deployment, monitoring, intent-based routing, source visibility, and AI-to-human escalation still need to be built and verified. The illustrative test above does not run an agent. Live use will also depend on Chatflow's AI provider and WhatsApp setup for each customer.

Start a conversation

Help shape how agents should work with your team.

Book a demo