Parlant: Build Custom AI Chat Agents for Your Website

What Is Parlant?

Parlant is an open platform for building customer-facing chat agents with AI. According to the project’s GitHub repository, you describe how the agent should behave, and your coding agent implements it for you. The platform provides an engine that keeps the agent on your rules in every conversation, plus the parts every customer-facing agent needs: a chat widget, human handoff, testing, hosting, and analytics.

Because the agent runs on your own code, it can use any system you can reach and any model you choose. Unlike closed customer experience platforms, you pay model providers directly instead of an overpriced fee per resolution. The project is licensed under Apache-2.0 and built in Python.

How It Works for Small Businesses

Imagine you run a small online store and need a chatbot that can answer order status questions. With Parlant, you write a simple Python script that defines a tool to look up orders, then tell the agent in plain language how to behave. For example, you can create a policy that says: “When a customer asks where an order is, ask for the order number if they haven’t given it, look it up with check_order_status, and give the status. Give delivery dates as estimates, never as promises.”

The platform includes a Dashboard where you can test sessions with simulated personas, check that the agent follows your rules, and then push changes live. You can also use pre-built scaffolds: run `parlant init my-agent –scaffold retail-store` for a working support agent, `–scaffold lead-gen` for a website agent that qualifies visitors and books meetings, or `–scaffold basic` for an empty one.

Why It Matters for Small Businesses

Customer-facing agents face pressures that assistant-style agents never do: a possibly confused, impatient, or hostile user. Parlant’s engine is designed to keep the agent on your rules consistently, even when a customer changes the subject, mixes different topics, or turns out to be on a different plan than they think.

The platform replaces the manual work of coding chat logic from scratch or paying per-resolution fees to closed platforms. You pay model providers directly—OpenAI, Anthropic, Gemini, LiteLLM, and custom providers are supported—instead of an overpriced fee per resolution. This can be a fraction of the cost of traditional customer experience platforms.

Getting Started

To get started, you install Parlant and set a key for the default OpenRouter provider. Then, in the Dashboard, create a Local development project and set its project token. You write a Python script that defines tools and policies, then run it. Back in the Dashboard, you can check the connection, open the project, and start testing sessions.

The quickstart gives you a prompt that has your coding agent install Parlant and set up an example agent or your own, one step at a time. From there, you can take it live on the Dashboard’s managed hosting and put it on your site with the chat widget. The project has 18.3k stars on GitHub and 1.6k forks, indicating an active community.

Key Benefits for Small Business Owners

Parlant allows you to build a custom AI chat agent without a full dev team. You describe the behavior in plain language, and the coding agent implements it. This means you can quickly add a chatbot to your website that handles FAQs, checks order status, or qualifies leads.

The platform is open-source and free to use, with no per-resolution fees. You only pay for the AI model usage directly to the provider. This can be significantly cheaper than closed customer experience platforms that charge per resolution.

Takeaways

• Parlant is an open-source platform for building customer-facing AI chat agents by describing behavior in plain language.
• It provides an engine that keeps the agent on your rules, plus a chat widget, human handoff, testing, hosting, and analytics.
• You pay model providers directly instead of per-resolution fees, making it cost-effective for small businesses.