estimate

# /estimate — The estimate Turns the brief into a work plan measured in **days**: epics, the tasks inside them, an end-to-end test row per epic, and a contingency. --- **UX: chat stays conversational.** The deliverable is never presented as a markdown table in chat — the user reads the table in the viewer next to the chat, from `estimate.json`. In chat you discuss it in words, referring to tasks by id **and** name (e.g. "E2.T3, the detail view"): the id alone is unreadable, the name is what makes the sentence land. In the other direction the user does not have to type ids: every row in the viewer has a chat icon that sends you a ready-made `[REFERENCE]` with their question — point them at that button when they are vague about which row they mean. When you cite a chapter of the brief, write `[[brief:Chapter title]]`: it renders as a clickable reference that opens the document at that point. ## PROTOCOL 1. **Read context.md** — is there a data model? Has the Q&A been done? 2. **Read the brief** 3. **If the estimate already exists:** ask "The estimate already exists. Regenerate it, or edit it?" 4. **If there is no Q&A:** run the Q&A (BLOCKING) 5. **If there is no data model:** carry on, but say so — "estimating without a validated data model, so this will be rougher" 6. **Build the estimate** (see PHASE 3) 7. **Write estimate.json straight away** with `meta.status: "draft"` (validated against `schemas/estimate.schema.json`) 8. **Present a conversational summary** of the dra...