> 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/news/november-2026.md). # November 2026 What changed in Dapta in November 2026: text agents that update your Dapta CRM after each conversation, dynamic CRM lists that update themselves, one Dapta CRM source for sequences with a live link to What reached Dapta in November 2026, newest first. Each change links to the page where it is documented in full. The month is still running, so new changes are added here as they ship. *** ## Text agents update your Dapta CRM after each conversation **A WhatsApp conversation now leaves its mark on the contact.** A text agent's **Conversation Analysis** tab has **Destinations**, with **Dapta CRM**. When a conversation closes, the agent: * finds the contact by phone, and creates it if you want, then fills the fields you choose with what the conversation captured; * can move the contact to a pipeline stage with rules you set, such as **Calificado** when `interested` equals `true`; * leaves a note on the contact, as a data list or free text. **Try it** runs the last closed conversation without moving or noting anything, and **History** shows what each conversation wrote. → [Post-Conversation Destinations](/ai-text-agents/setting-up-your-text-agent/post-conversation-destinations.md) *** ## Dynamic lists that update themselves **Say who belongs, and the list keeps itself up to date.** **New list** in **CRM** → **Lists** now has **Static** and **Dynamic**. A dynamic list holds every contact that matches its conditions: name, phone, creation date, tags or your own fields. * **And / or.** Conditions in a box must all match; **Add alternative** adds another box, and matching any box is enough. * **See who matches before you save.** The editor shows how many contacts match and a sample. * **Always current.** When a contact's data changes, it joins or leaves the list on its own. **Edit rules** changes the conditions any time. * **Ready for sequences and pipelines** once it has finished filling. Up to 20 dynamic lists per workspace. → [Dynamic lists](/crm/crm-lists.md#dynamic-lists) *** ## One Dapta CRM source for your sequences **Start a voice or WhatsApp sequence from the contacts you already have.** Step 3 of the sequence wizard now has a single **Dapta CRM** source with three choices: * **All contacts** or **A list**: a one-time import of the contacts in your CRM, or in one of your [contact lists](/crm/crm-lists.md), at that moment. Contacts added later are not picked up. * **A pipeline**: whoever is in a stage, with a **Live sequence** switch. On, the stage stays linked to the sequence, so whoever enters the stage later is enrolled too. Off, the contacts in the stage are enrolled once. * **Clearer steps.** The wizard shows a dot per step instead of tabs, and a voice sequence fed by a pipeline now reaches **Property mapping** with the stage's CRM fields. → [Create a voice sequence](/ai-voice-agents/how-to-create-your-first-campaign-in-dapta/create-a-campaign.md#use-your-dapta-crm) · [Create a text sequence](/ai-text-agents/campaigns/create-a-campaign.md#3-choose-where-the-contacts-come-from) *** ## Do Not Call from the contact **An open CRM contact shows whether its number can be called, and changes it in one click.** The phone icon in the contact's header is red while the number is on the workspace's Do Not Call list. Click it to add the number; removing it asks you to confirm. The red **Do Not Call** tag next to the name says who added it, when and why, with a link to the call when a voice agent added it. → [Do Not Call from a contact](/crm.md#do-not-call-from-a-contact) · [Do Not Call list](/ai-voice-agents/how-to-create-your-first-campaign-in-dapta/do-not-call-list.md#add-or-remove-a-number-from-a-contact) *** ## Your CRM from Claude, ChatGPT and other AI assistants **The Dapta MCP server now reaches your CRM.** From Claude, ChatGPT, Cursor or any assistant connected to your workspace, you can: * **Read**: search contacts, open a contact and its timeline, and read your lists, pipelines and boards. * **Change**: create or update a contact, add a note, move contacts to a stage, and add contacts to a list or a pipeline. * **Review every change first.** Each change is shown as a preview you confirm before anything is saved. Moving contacts to a stage that is linked to a sequence enrolls them in it, so check a stage's linked sequences before you move people into it. → [CRM through the MCP server](/dapta-mcp/skills/crm.md) · [Tools reference](/dapta-mcp/tools-reference.md) *** ## Contacts from HubSpot or Google Sheets, kept up to date **Bring the list your team keeps in HubSpot, or a Google Sheet, into the CRM without a file.** **Upload contacts** in the CRM now asks where the contacts come from: a file, a HubSpot list or a Google Sheet. * **Map once.** Match each HubSpot property or sheet column to a CRM field, a new column or a pipeline stage. Invalid numbers are left out and listed for you; they don't stop the import. * **Kept up to date.** **Keep up to date** is on by default: every 15 minutes Dapta brings the new or changed contacts. Turn it off for a one-time import. * **Connected sources.** A sync icon in the toolbar lists every source with its last run, so you can sync now, pause, resume or disconnect it. * **Nothing is deleted.** A contact who leaves the list, or a row you delete, stays in your CRM. → [Bring contacts from HubSpot or Google Sheets](/crm.md#bring-contacts-from-hubspot-or-google-sheets) · [Connected sources](/crm.md#connected-sources) *** ## Today's models in Flow Studio AI nodes **The OpenAI, Claude and Google Gemini nodes offer each lab's current models.** New nodes start on GPT-5.6 Terra, Claude Sonnet 5.5 or Gemini 3.5 Flash-Lite, and the menus add models such as GPT-6.1 Sol, Claude Fable 5.1 and Gemini 4 Argon. Older models leave the menu, but nodes that already use them keep working. → [AI nodes](/flow-studio/ai-nodes.md#the-models) *** ## Reports Dapti builds and you can share **Ask Dapti for a report and it builds one from your data**, with figures, charts and tables, as a card you open beside the chat. * **Every version is kept.** Ask for a change and Dapti saves a new version; earlier ones stay one click away. * **Share it.** Keep it for your workspace members, or publish a link anyone can open. A public link always shows the latest version and stops working as soon as you switch back to members only. → [Reports and files in Dapti](/dapti/capabilities/reports-and-files.md) *** ## Files you attach to Dapti, and tables as CSV **Dapti reads the files you give it.** Attach Excel (`.xlsx`), CSV and TSV sheets, PDFs, Word documents, images, text and code: pick them, paste them or drop them into the chat, up to 10 per message. Each file shows its progress, and one Dapti can't take tells you why. **Ask for a table as a file** and it arrives as a CSV download card, like Dapti's PDF and Word files, with its number of rows and columns. It works for a table already in the conversation or a file you attached, up to 20,000 rows. → [Work with your own files](/dapti/capabilities/reports-and-files.md#work-with-your-own-files) · [Turn a table into a CSV](/dapti/capabilities/reports-and-files.md#turn-a-table-into-a-csv) --- # 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/news/november-2026.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.