What does this skill do?

The Skill Product Capabilities Plan translates product intent into explicit engineering constraints, creating a reusable artifact that persists across different environments and sessions. Starting from a PRD, roadmap item, or product note, it generates an SRS-style implementation contract with actors, surfaces, states, interfaces, and observability requirements before coding begins.

PRD with implicit restrictions
When there is a PRD or product specification, but the implementation restrictions remain vague and need to be clarified.
Multi-service functionality
When a resource spans multiple services, repositories, or teams and requires a resource contract before scheduling.
New Integrations
When you request to add an integration (e.g., Stripe) and need to expose idempotent webhooks, handle errors, and distinguish between promises and fulfillment.
Reusable item
When senior engineers keep making the same unstated assumptions during reviews, and you need a document that doesn't rely on scattered notes or chat context.

Usage examples

📄 PRD on a contract basis
I have this PRD to allow a client's synchronization to be paused without affecting other tenants. It generates the complete capacity plan.
🔌 New integration
The product team wants to add Stripe as a payment gateway. Create the implementation contract with idempotent webhooks and error handling.
🔗 Multi-service
I need a requirements specification for a feature that involves the auth, billing, and notifications services before I start coding.
♻️ Architecture Review
Turn this product note into a reusable SRS artifact with constraints, invariants, and a clear handoff for the engineering team.

Features

Capacity Reformulation Summarize the request in a concise statement: who the user is, what new capability will be available, and what outcome will change.
Resolving Restrictions Extract business rules, scope limits, invariants, confidence limits, data ownership, and lifecycle transitions.
SRS Implementation Contract It produces actors, surfaces, required states, interfaces, data model implications, and observability requirements.
Non-goals and open questions Explicitly define what the workflow does not cover and list any blockers or pending product decisions.
Structured Handoff Indicate whether it is ready for implementation, needs an architectural review, or requires product clarification, along with the suggested next steps.

Frequently asked questions

It accepts PRDs, issues, roadmap items, or product notes. It also leverages the current architecture (repositories, schemas), existing RFCs, and delivery constraints such as authentication or billing.
It only generates the planning artifact (an SRS-style capacity contract). It does not write code; the handoff specifies which subsequent workflow it should follow, such as tdd-workflow.
Yes. It is designed specifically for multi-service or multi-repository scenarios where a shared capacity agreement is required before scheduling.
Yes. The goal is to produce a durable document that does not rely on scattered notes or messaging context, and that can be reused across different AI assistants and environments.
Product Capabilities Plan — Engineering Constraints with Claude AI

¿Prefieres escuchar el contenido? Genera la narración de audio con un clic.