ai-feature-prd

# AI Feature PRD Skill AI features break the normal PRD because the system is probabilistic: it will be wrong sometimes, and the product must be designed around that, not in denial of it. This skill extends a standard PRD with the AI-specific sections that decide whether the feature is trustworthy — the UX of uncertainty, the eval bar, guardrails, and what happens when the model is wrong. ## Required Inputs Ask for these only if they aren't already provided: - **The user problem** and why an AI/probabilistic approach fits it (vs. deterministic rules). - **What "good" looks like** to the user, and the cost of a wrong answer (low-stakes vs. high-stakes). - **Inputs available** — context/data the model can use; privacy constraints. - **Trust level needed** — can the user verify the output, or must it be near-perfect? ## Output Format ### AI Feature PRD: [feature] **1. Problem & why AI** — the user problem, and why a model (not rules) is the right tool. If rules would do, say so. **2. Experience** — the core flow, and crucially the **UX of uncertainty**: how confidence is shown, how the user verifies/edits, and how errors are made cheap to recover from. AI features live or die here. **3. Model approach** — prompt / fine-tune / RAG / agent (link [`rag-design-doc`](../rag-design-doc/SKILL.md) or [`agent-spec`](../agent-spec/SKILL.md)), the model tier, and why. **4. Quality bar & evaluation** — the metrics and the explicit ship threshold; reference an [`ai-eval-plan`](../a...