canonical: https://jentic.com/apis/chat-data.com/chat-data # Chat Data API Schema Guide The Chat Data API builds AI chatbots trained on your own documents, website links, and Q&As, then serves their answers over chat, live chat, and WhatsApp. It queries a chatbot's knowledge base, scores answer confidence, and records conversations with attached leads and customers. Chatbot settings, white-label options, webhooks, and sub-users round out the platform. ## For AI agents Create and train custom AI chatbots on your data, send and receive chat messages, query the knowledge base, and manage conversations through the Chat Data API. ## Scope Does not handle raw language-model hosting, voice calls, or email campaigns. Use for building and serving AI chatbots trained on your own data only. ## Capabilities - Create and train an AI chatbot on documents, links, and Q&As - Send a message to a chatbot and receive its grounded reply - Query a chatbot's knowledge base and score answer confidence - Retrain a chatbot after its source material changes - Read conversations and attach leads and customers to them - Run live chat and send outbound WhatsApp messages - Configure chatbot settings, white-label, webhooks, and sub-users ## Use cases ### AI Agent Chatbot Orchestration An AI agent connected through Jentic can stand up a support bot end to end: it creates a chatbot from a set of document URLs, checks the training status until it is ready, and sends a test message to confirm the bot answers. Chat Data handles ingestion and retrieval so the agent orchestrates setup rather than wiring each step. Example prompt: Create a chatbot from a list of document URLs, poll its training status until ready, and send a test message to confirm it answers ### Support Chatbot on Your Docs Answer customer questions from your own content by training a chatbot on documents, website links, and Q&As. Chat Data grounds replies in that knowledge base and returns a confidence score, so a workflow can escalate low-confidence answers to a human. Example prompt: Query the chatbot's knowledge base for a customer question and return the answer with its confidence score ### Omnichannel Live Chat Hand off from bot to human and reach users on messaging channels by running live chat and sending outbound WhatsApp messages. Chat Data toggles bot activation per conversation so a workflow can pause the bot when an agent takes over. Example prompt: Pause the bot in a conversation, send a live-chat reply, then send an outbound WhatsApp message to the user ### Conversation and Lead Capture Turn chats into pipeline by reading conversations and attaching leads and customers to them. Chat Data exposes conversation history and lead attachment so captured contacts flow to downstream systems. Example prompt: Retrieve a conversation by its identifier and attach a lead with the captured contact details ## Key endpoints | Method | Path | Description | | --- | --- | --- | | POST | `/create-chatbot` | Create and train a chatbot | | POST | `/chat` | Send a message to a chatbot | | POST | `/query-knowledge-base/{chatbotId}` | Query the knowledge base | | GET | `/get-chatbots` | List all chatbots | | POST | `/retrain-chatbot` | Retrain an existing chatbot | | GET | `/get-conversations/{chatbotId}` | Get conversations for a chatbot | | POST | `/add-webhook` | Add a webhook | ## Key resources - **Chatbots** — Create, train, retrain, retrieve, update, and delete chatbots - **Knowledge Base** — Query the knowledge base, score confidence, and manage uploaded files - **Conversations** — Read conversations, attach leads, and manage customers - **Live Chat** — Live-chat messaging, bot control, and WhatsApp outbound - **Webhooks** — Add and delete chatbot webhooks - **Sub-users** — Create, list, update, and delete sub-users ## AI readiness This API is usable in Jentic One now. Its AI-readiness score against Jentic's framework shows where it stands today and where improvements would make it even easier for agents to use. - **Score:** 65 / 100 - **Maturity:** AI-Aware - **Dimensions:** - Foundational Compliance: 59 / 100 - Developer Experience & Jentic Compatibility: 71 / 100 - AI-Readiness & Agent Experience: 51 / 100 - Agent Usability: 94 / 100 - Security: 60 / 100 - AI Discoverability: 80 / 100 - **View full report:** https://jentic.com/apis/chat-data.com/chat-data/scorecard - **How the score is calculated:** https://docs.jentic.com/reference/api-readiness-framework/overview/ - **More about the dimensions:** https://docs.jentic.com/reference/api-readiness-framework/specification/#dimensional-model-overview ### Score it yourself Every API in the directory is allowlisted, so you can re-score it with no key required. - **Score your own API:** https://jentic.com/scorecard.md - **Scoring CLI agent skill:** https://github.com/jentic/jentic-api-scorecard/blob/main/skills/jentic-api-scorecard/SKILL.md ```sh npx @jentic/api-scorecard-cli score ``` ## Why Jentic - **Setup:** Wiring Chat Data by hand means handling its bearer-token auth and tracking the chatbot, knowledge-base, conversation, and live-chat paths yourself. Through Jentic you install once, import Chat Data from the API Directory, store the token once, and your agent calls it. - **Permission scoping:** You choose which Chat Data operations the agent may call, such as sending messages and querying the knowledge base, so operations like deleting chatbots or conversations are not included unless you add them. - **Credential handling:** Your Chat Data token is stored once, encrypted, by your own Jentic One instance and injected at execution time. It never enters the agent's prompt, logs, or context. - **Discovery method:** Agents search Jentic by intent such as 'send a chatbot message' or 'query the knowledge base', and Jentic returns the matching Chat Data operation with its input schema so the agent calls the right endpoint without browsing the reference docs. ## Related APIs - **ChatNode API** — ChatNode builds trained chat agents and serves their conversations. - **OpenAI API** — OpenAI provides raw language models for generation and reasoning. - **Cohere API** — Cohere offers language models and embeddings for retrieval and classification. ## FAQ ### What authentication does the Chat Data API use? The Chat Data API authenticates with a bearer token sent in the Authorization header, per its OpenAPI spec. You create the token in your Chat Data account, and through Jentic it is stored encrypted by your own Jentic One instance and injected at call time, so it never appears in the agent's prompt or logs. ### Can I train a chatbot on my own data with the Chat Data API? Yes. You create a chatbot from documents, website links, plain text, or Q&As, then retrain it when the sources change. The API reports training status and lets you query the resulting knowledge base with a confidence score on each answer. ### Can I run live chat and WhatsApp with the Chat Data API? Yes. The API sends and receives chat messages, runs live chat with bot control per conversation, and sends outbound WhatsApp messages. You can pause or resume the bot in a conversation when a human agent takes over. ### Is there a Chat Data MCP server? You do not need an MCP server to give your agent the Chat Data API. Jentic connects it directly from the API Directory: import Chat Data, store your token once, and your agent creates chatbots and answers messages through the operations the API already exposes. Operations are discovered on demand, so no extra tool definitions are loaded into the agent's context. ### Can I limit what my agent is allowed to do with the Chat Data API? Yes. Write a rule that allows only the operations the agent needs, for example sending messages and querying the knowledge base, and leave out operations like deleting chatbots or conversations. The agent serves answers within those bounds, and every call it makes is logged. ### How do I create a chatbot through Jentic with the Chat Data API? Search Jentic for 'create and train a chatbot', which returns the Chat Data operation with its input schema, then supply the source documents and settings and the agent runs it. To run it on your own infrastructure, install Jentic One from its GitHub repo.