- Home
- Skills
- Martinffx
- Claude Code Atelier
- Atelier Spec Product
atelier-spec-product_skill
- JavaScript
4
GitHub Stars
1
Bundled Files
2 months ago
Catalog Refreshed
4 months ago
First Indexed
Readme & install
Copy the install command, review bundled files from the catalogue, and read any extended description pulled from the listing source.
Installation
Preview and clipboard use veilstrat where the catalogue uses aiagentskills.
npx veilstrat add skill martinffx/claude-code-atelier --skill atelier-spec-product- SKILL.md5.0 KB
Overview
This skill guides requirements discovery and scope definition for feature specifications. It helps run discovery interviews, turn insights into user stories, prioritize work, and prepare clear handoffs to architects. Use it to create focused MVPs and reduce risk before implementation.
How this skill works
Run structured discovery using open-ended prompts to surface problem statements, user needs, constraints, and success metrics. Convert findings into scoped in-scope/out-of-scope definitions, MVP criteria, and decomposed user stories with acceptance criteria. Apply prioritization frameworks and produce deliverables that feed directly into technical design.
When to use it
- At project kickoff to discover and validate the problem space
- When conducting stakeholder or user discovery interviews
- Before scoping work for an MVP or release
- To prioritize features and reduce delivery risk
- When preparing product outputs for architect handoff
Best practices
- Ask open-ended questions to reveal real user jobs and pain points
- Separate core MVP functionality from nice-to-haves explicitly
- Write user stories in the As a/I want/So that format with clear acceptance criteria
- Prioritize using value-vs-effort and risk-first sequencing
- Document integration points and non-functional requirements for architects
Example use cases
- Discovery interview to validate a new onboarding flow and define the MVP
- Scope definition for a payment integration, listing in-scope APIs and out-of-scope features
- Decomposing a large reporting feature into independent stories ordered by risk
- Prioritizing backlog items using MoSCoW and value-effort mapping
- Preparing a product packet (stories, data needs, constraints) for the architecture team
FAQ
Defer features that are not required for delivering the primary user value or that introduce high effort without validated benefit. Mark them as future iterations with criteria for reconsideration.
What makes a good MVP definition?
A good MVP delivers the smallest set of functionality that lets users accomplish the core job, validates key assumptions, and yields measurable learning for the next iteration.