Feature Intake Brief
Turn a rough feature request into a one-page brief with RICE scoring against your backlog and a build now/later/validate/no call.
Before & after
Same prompt, same agent — the only difference is the skill.
Adding AI-powered search is a feature a lot of customers are asking for these days. You'd want to make sure it actually solves a real problem for your users though. Some things to think about: do users actually search your knowledge base, or do they mostly browse? How often do they search? What kinds of queries do they run? You might want to talk to some customers to understand their pain points better. There are different approaches you could take — embedding-based search, semantic search, that kind of thing. You should research the options and think about what makes sense for your product and your technical resources. The user experience would be important too. What would the search interface look like? How would results be ranked and displayed? Let me know if you want to brainstorm more on this.
PRODUCT STRATEGY: AI-POWERED KNOWLEDGE BASE SEARCH PROBLEM HYPOTHESIS Current category search (keyword matching) yields 23% no-results queries (internal telemetry). Users resort to browsing or giving up entirely. Semantic search with ranked relevance could solve for query intent instead of exact matches. RESEARCH GATES • Interview 12 power users: How often do you search? Do results match intent? Where do you go when search fails? • Analyze query logs: High-volume mismatches (queries returning 0 results, users immediately re-query with different terms). • Competitive audit: How do Slack, Confluence, and Linear rank semantic results? What UX patterns drive engagement? SCOPED MVP → Embed existing knowledge base content (300-5000 articles depending on size) → Semantic ranking layer only; keep keyword search as fallback → Expose relevance score; mark AI-reranked results visually → Measurement: Compare abandon rate (no-result queries) before/after; track CTR on top 3 results DEFER: Question rephrasing, personalized ranking, result summaries. Test core hypothesis first. RISKS: Hallucinated answers if knowledge base has gaps; cost of embeddings at scale; customer complacency (they'll want summaries immediately).
About this skill
Ideas arrive half-formed and loud. This skill turns one idea into an honest one-page brief, finds the smallest version worth building, and places it against your current backlog so the decision is explicit.
What it does
- Rewrites the request as a problem statement and considers alternatives, including non-product fixes
- Rates the strength of the evidence and names the gaps
- Defines the smallest viable scope with a t-shirt size and the biggest unknown
- Scores RICE with a source for each input and shows where the idea ranks in your backlog
What you get
A feature brief with the problem, evidence, options, scope, RICE score and backlog position, a decision (build now, later, validate first, or no), the trigger to revisit, and a draft reply to the requester.
Best for
Product managers and founders triaging requests from sales, customers, and executives.
SKILL.md preview
---
name: feature-intake-brief
description: Turn a rough feature idea or request into a one-page feature brief - problem, evidence, target user, smallest viable scope, rough size - score it with RICE against the current backlog, and return a build now / later / no decision. Use when someone brings a feature idea ("customers keep asking for X", "the CEO wants Y", "should we build Z"), when triaging requests, or when the user needs to decide whether an idea deserves a full spec, including casual asks like "is this worth building", "sales keeps asking for X, thoughts?", or "where would this rank on our backlog".
---
# Feature Intake Brief
… (sign up to view the full skill)More business & ops skills
View all Business & Ops skills →Product Practice Diagnostic
Trace a product-team symptom (unused features, roadmap churn, priority fights) to the broken stage of the loop and fix one thing first.
Feature Launch Go-to-Market Plan
Plan one feature launch: tier, plan access, audience messages, readiness gates for support, sales, billing, and docs, and success metrics.
Pricing Tier and Packaging Diagnosis
Finds why your pricing tiers underperform and redesigns value metric, fences, price gaps, and upgrade triggers with a migration plan.
Product Specs, Roadmaps, and Research Decisions
Turn customer evidence into testable PRDs, dependency-aware roadmaps, and stakeholder updates with clear decisions and tradeoffs.