# Generated by API Evangelist (build-phrasing.py). Our phrasing, not observed demand. overlay: 1.0.0 info: title: API Evangelist conversational phrasing for TextQL v2 Chat API version: 1.0.0 extends: openapi/textql-chat-api-openapi.yml actions: - target: $.info update: x-apievangelist-phrasing: method: generated generated: '2026-10-01' generator: build-phrasing.py label: Generated by API Evangelist operations: 8 - target: $.paths['/v2/chats'].get update: x-apievangelist-phrasing: intent: List my API chats effect: read questions: - Which chats has my API key created? - Can I search my chats by summary or first message? instructions: - text: List my chats. - text: Find chats mentioning {search_term}, {limit} at a time. slots: search_term: query.search_term limit: query.limit method: generated generated: '2026-10-01' - target: $.paths['/v2/chats'].post update: x-apievangelist-phrasing: intent: Ask a data question and wait for the answer effect: write questions: - How do I ask TextQL a question about my data and get the full answer back in one response? - Can I attach a file when I ask a question? - Can I limit which connectors a question is answered from? instructions: - text: Ask {question} and wait for the complete answer. slots: question: requestBody.question - text: Ask {question} using only connectors {connectors}, returning the answer synchronously. slots: question: requestBody.question connectors: requestBody.connector_ids - text: Continue chat {chat} with the follow-up {question} and return the whole reply at once. slots: chat: requestBody.chat_id question: requestBody.question method: generated generated: '2026-10-01' - target: $.paths['/v2/chats/stream'].post update: x-apievangelist-phrasing: intent: Ask a data question and stream the answer effect: write questions: - Can I get the answer to a data question streamed back as it's generated? - How do I receive Server-Sent Events with text deltas and execution cells for a question? instructions: - text: Stream the answer to {question} as server-sent events. slots: question: requestBody.question - text: Stream a reply to {question} in chat {chat} using model {model}. slots: question: requestBody.question chat: requestBody.chat_id model: requestBody.model method: generated generated: '2026-10-01' - target: $.paths['/v2/chats/{id}'].get update: x-apievangelist-phrasing: intent: Get a chat with its messages effect: read questions: - How do I retrieve a chat along with its messages and generated assets? - What did the agent answer in a previous REST chat? instructions: - text: Get chat {id} with its messages. slots: id: path.id - text: Show the conversation and assets for chat {id}. slots: id: path.id method: generated generated: '2026-10-01' - target: $.paths['/v2/chats/{id}/cells'].get update: x-apievangelist-phrasing: intent: List the execution cells of a chat effect: read questions: - Can I see the SQL and Python the agent ran to produce an answer? - How do I page through every step behind a chat's answers? instructions: - text: List the cells of chat {id}. slots: id: path.id - text: Show {limit} execution steps from chat {id} starting at offset {offset}. slots: limit: query.limit id: path.id offset: query.offset method: generated generated: '2026-10-01' - target: $.paths['/v2/chats/{id}/cells/{cellId}'].get update: x-apievangelist-phrasing: intent: Get one execution cell from a chat effect: read questions: - How do I fetch a single step from a chat by its cell ID? - Can I look at just one generated query and its output? instructions: - text: Get cell {cell} from chat {id}. slots: cell: path.cellId id: path.id - text: Show the output of step {cell} in chat {id}. slots: cell: path.cellId id: path.id method: generated generated: '2026-10-01' - target: $.paths['/v2/chats/{id}/cancel'].post update: x-apievangelist-phrasing: intent: Cancel a streaming chat response effect: destructive questions: - How do I stop a REST chat stream that's taking too long? - Can I cancel an in-flight streamed answer? instructions: - text: Cancel the running stream for chat {id}. slots: id: path.id - text: Abort the streamed answer currently generating in chat {id}. slots: id: path.id method: generated generated: '2026-10-01' - target: $.paths['/v2/models'].get update: x-apievangelist-phrasing: intent: List models I can run chats on effect: read questions: - Which models is my organization allowed to use for chats? - What model IDs can I pass when asking a question? instructions: - text: List the models available to me for chats. - text: Show which chat models my role is allowed to pick. method: generated generated: '2026-10-01'