> For the complete documentation index, see [llms.txt](https://docs.dapta.ai/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.dapta.ai/dapta-mcp/skills/text-agents.md). # Text & WhatsApp Agents Build text and chat agents with Claude, test and tune how they respond, manage their versions, and connect them to WhatsApp, all from a plain conversation. Once Dapta is connected to Claude, you can build text and chat agents and get them ready for your website, WhatsApp, and other channels, just by asking. {% hint style="info" %} Anything that changes an agent is done as a **preview, then confirm**. Claude shows you the exact change and waits for your yes before it applies it. Reading and looking things up never needs confirmation. {% endhint %} ## Build a text or chat agent Ask Claude to create a new agent and it walks you through the few things it needs: what the agent is for, what language it should speak, and your company details. {% hint style="info" %} **Creating an agent from a website.** If you share your website, Claude reads it with the web tools of the client you are in (claude.ai, Claude Desktop and Claude Code have them). If your client has no web access, Claude asks you for the company details instead. {% endhint %} **Try saying:** * "Create a customer service chat agent for my company." * "Set up a sales chat agent, here's our website: example.com" * "I want a receptionist chat agent in Spanish for my dental clinic." You get back a preview of the agent, and after you confirm, a ready-to-use agent with your company information built in and a few suggested starter questions your website chat can offer visitors. **Behind the scenes:** `get_workspace_company_context`, `preview_create_text_agent`, `commit_create_text_agent`, `generate_text_agent_example_questions`. *** ## Test and tune how an agent responds If an agent is answering something wrong, using the wrong tone, or missing a handover, ask Claude to fix it. Claude reads the agent's current setup, tests it or looks at recent real conversations to spot the problem, and proposes a focused change in plain language. You approve each change one at a time. **Try saying:** * "Why is my agent answering pricing questions like that?" * "Send my chat agent a test message about refunds and show me the answer." * "Rewrite my support agent's instructions to be more concise and friendly." You get a clear before-and-after preview of every change. You can keep looping with Claude to test and tune until the agent behaves the way you want. {% hint style="warning" %} Some older agents cannot be edited here. If an agent was built in an earlier version of Dapta, Claude tells you plainly and points you to where you can update it instead. {% endhint %} **Behind the scenes:** `get_text_agent_details`, `get_text_agent_conversations`, `list_conversations`, `get_conversation`, `chat_with_text_agent`, `preview_update_text_agent`, `commit_update_text_agent`. *** ## Manage versions Ask Claude to read an agent's history, publish or discard a pending draft, roll back to an earlier version, edit the shared identity context behind it, or delete an agent nobody uses. After each change, Claude re-reads what is live to confirm it. **Try saying:** * "Show me the version history of my support agent." * "Publish the draft of my sales agent." * "Roll my agent back to the previous version." **Behind the scenes:** `get_text_agent_history`, `preview_publish_text_agent` and `commit_publish_text_agent`, `discard_text_agent_draft`, `preview_rollback_text_agent` and `commit_rollback_text_agent`, `preview_update_text_agent_context` and `commit_update_text_agent_context`, `preview_delete_text_agent` and `commit_delete_text_agent`. *** ## Inspect an agent before changing it You can also just look. Ask Claude to show an agent's current instructions, the company information it knows, or its persona and language, without touching anything. **Try saying:** * "Show me my agent's instructions." * "What does my chat agent know about my company?" * "What language and persona is my support agent set up with?" **Behind the scenes:** `get_text_agent_details`. *** ## Find and compare your agents Ask Claude to list your agents, find a specific one, or see which ones are busiest and which channels they are connected to. **Try saying:** * "What chat agents do I have?" * "Which of my agents handles the most messages?" * "What WhatsApp lines are connected to my agents?" You get back a clear summary, rankings by message volume, and full details on any agent you name. **Behind the scenes:** `list_all_text_agents`, `search_text_agent`, `top_agents_by_messages`, `get_workspace_whatsapp_lines`. *** ## Connect an agent to WhatsApp Claude helps you see your WhatsApp business lines, check whether an agent already has WhatsApp connected, find which agent owns a number, and understand exactly what you need before you finish connecting a new number. **Try saying:** * "Which WhatsApp lines does my workspace have?" * "Is my Soporte agent connected to WhatsApp, and to which number?" * "I want to hook a new WhatsApp number up to my Ventas agent, what do I need?" {% hint style="info" %} Claude can prepare everything and tell you exactly what is required, but the final step of registering a new WhatsApp number is completed inside Dapta, on the agent's channels page. {% endhint %} **Behind the scenes:** `get_workspace_whatsapp_lines`, `get_text_agent_details`, `search_text_agent`, `list_all_text_agents`, `preview_update_text_agent`, `commit_update_text_agent`. --- # Agent Instructions This documentation is published with GitBook. GitBook is the documentation platform designed so that both humans and AI agents can read, navigate, and reason over technical content effectively. Learn more at gitbook.com. ## Querying This Documentation If you need additional information that is not directly available in this page, you can query the documentation by asking a question. Perform an HTTP GET request on the following URL with the `ask` and `goal` query parameters: ``` GET https://docs.dapta.ai/dapta-mcp/skills/text-agents.md?ask=&goal= ``` `ask` is the immediate question: it should be specific, self-contained, and written in natural language. `goal` is what the user is ultimately trying to achieve, the reason they need the answer. Sharing it helps GitBook give you a better, more relevant answer. A goal is most helpful when it describes the outcome the user wants rather than restating the question. For example, with `ask=how do I create an API token`, a goal like `automate deployments from our CI pipeline` lets GitBook tailor the answer to that use case. The response will contain a direct answer to the question and relevant excerpts and sources from the documentation. Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.