Sharpener.dev¶
Sharpener is an AI-powered product scoping tool that transforms a rough idea into a complete, ready-to-build Product Requirements Document (PRD).
Overview¶
In the AI era, AI is only as smart as your plan. Sharpener addresses this by applying rigorous product thinking to the front end of development — before a single line of code is written.
The tool is built on the BMAD Method framework, a product management methodology that surfaces questions about anything the framework isn't 100% sure on. This ensures the resulting PRD is genuinely comprehensive, not just superficially complete.
The Problem It Solves¶
Most product ideas start as vague one-liners: "an app that tells you what to cook from a photo of your fridge." The gap between that sentence and a buildable spec is enormous — and most teams either skip the gap (jumping straight to coding) or drown in it (months of planning before progress).
Sharpener collapses that gap with a guided, AI-facilitated process that takes a rough idea and produces a PRD that's:
- Structured enough to be actionable
- Flexible enough to handle ambiguity
- Ready to paste directly into Claude Code, Cursor, or Codex
How It Works¶
Sharpener uses a 3-step process:
- Rough Idea — Start with a single sentence describing the product
- Cover Every Angle — The BMAD Method walks through 9 structured sections, surfacing questions about unknowns
- Hand Off Ready-to-Ship PRD — Output is markdown formatted for direct use in AI coding tools
The BMAD 9 Sections¶
The core of Sharpener is the BMAD Method's 9-section PRD structure. This same structure appears in the Kelly factory's pipeline-stages as the Stage 3 (Planning) output, where quality-gates validate the PRD before implementation begins.
| # | Section | What It Covers |
|---|---|---|
| 1 | Discovery | What problem does this solve? Who has it? |
| 2 | Success Criteria | How do we know it worked? (User, Business, Technical) |
| 3 | User Journeys | Who uses it and how? Happy paths + edge cases |
| 4 | Domain | What is the product category? What are the rules? |
| 5 | Innovation | What's novel? What assumptions must we validate? |
| 6 | Product Type | Mobile app? Web? API? Library? |
| 7 | Scope | MVP vs. future features — what to ship vs. defer |
| 8 | Functional Requirements | What must the system do? |
| 9 | Non-Functional Requirements | Performance, security, accessibility, etc. |
Key UX Wedge: Photo-First Input¶
Sharpener's demo product ("Plantry") showcases the camera-to-recipe flow as the core UX insight — a single photo becomes the canonical source of truth for available ingredients, collapsing the hardest part of meal planning.
This illustrates Sharpener's philosophy: photo-first input as a UX wedge. The interaction users struggle most with (manually typing ingredients) is replaced by the most natural action (taking a picture).
Key Product Decisions Demonstrated¶
The Plantry demo illustrates several important product thinking patterns:
Confidence-Driven UX¶
- Per-item confidence scores for detected ingredients
- Ingredient Confirmation modal for low-confidence matches: Confirm / Suggest Substitutes / Manual Edit
- No free-form text entry that could introduce errors — the photo is ground truth
Strict Visibility Constraint (MVP Posture)¶
MVP matches recipes only to ingredients visible in the photo. This:
- Sets a strict, testable boundary for the matching model
- Increases predictability (no surprise shopping trips)
- Reduces cognitive overhead from inferred/assumed ingredients
Build Priority: Camera Capture First¶
Sharpener recommends starting with the camera capture flow and image upload pipeline — highest risk, unlocks two critical paths. Without a working photo→ingredients flow, nothing else matters.
Solo-Builder Philosophy¶
Sharpener is designed for solo builders who want managed services over custom backend plumbing. The PRD output is optimized for direct use in AI coding tools — not for handoff to a large team with a traditional PM org.
Pricing¶
The primary CTA is "Sharpen my idea" — routes to a pricing section. The tool is positioned as accessible to independent builders, not just enterprise teams.
Relationship to BMAD Method¶
Sharpener is a practical application of the BMAD Method (github.com/bmad-code-org/BMAD-METHOD). The 9-section PRD structure directly mirrors BMAD's elicitation framework.
Sharpener's output is directly usable as the prd.md artifact in the Kelly factory's full-pipeline Stage 3 (Planning). The 9-section structure aligns with the factory's PRD requirements, making sharpener.dev a potential upstream tool for factory intake — a solo builder could use Sharpener to scope an idea, then feed the resulting PRD directly into the factory pipeline.
Cross-reference: See bmad-method.md for the full methodology documentation.
See Also¶
- BMAD Method Documentation
- Plantry PRD Demo (example output)
- BMAD Method — the framework underpinning Sharpener
- software-factory — the pipeline Sharpener's output feeds into
- full-pipeline — Stage 3 Planning where Sharpener's PRD is consumed
- quality-gates — validates PRD completeness before implementation begins
- tea-audit — tests Non-Functional Requirements from Section 9
- discovery-gate — Discovery Gate (28/40) score validates opportunity before planning begins; Sharpener output provides the structured PRD that clears Stage 3 gate criteria
Bibliography¶
- Sharpener.dev — https://sharpener.dev/
- BMAD Method Documentation — https://docs.bmad-method.org
- BMAD Method GitHub — https://github.com/bmad-code-org/BMAD-METHOD
Source Attribution¶
Sharpener.dev — https://sharpener.dev/