feature-spec

Generate detailed feature specifications through codebase exploration and structured user discussion.

Updated Apr 13, 2026
One-click install
npx skills add https://github.com/tsmes/conaro --skill feature-spec-tsmes
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: feature-spec
Source: https://github.com/tsmes/conaro/tree/main/.claude/skills/feature-spec
Command: npx skills add https://github.com/tsmes/conaro --skill feature-spec-tsmes

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? Writing feature specifications often suffers from ambiguity, scope creep, and misalignment between stakeholders. This Skill eliminates vague requirements by exploring the codebase first, asking focused one-at-a-time questions, and converging on a verifiable specification before any implementation begins. ## Core Features & Use Cases - Codebase-Informed Clarification: Investigates existing patterns, conventions, and constraints using parallel sub-agents before asking the user anything, so questions are grounded in reality. - Structured Discussion Workflow: Follows a five-step workflow (Investigate, Clarify, Explore approaches, Align, Specify) with one question per message and multiple-choice options to reduce cognitive load. - Verifiable Output Format: Produces a specification with problem statement, numbered requirements, in/out of scope lists, acceptance criteria, constraints, and open questions, saved to .claude/plans/. - Use Case: A developer needs to add a notification feature to a web app. The Skill explores the existing architecture, asks targeted scoping questions with recommendations, proposes 2-3 approaches with trade-offs, and writes a complete spec ready for implementation planning. ## Quick Start Ask the AI to create a feature specification for the new feature you have in mind and let it explore the codebase first.

Frequently Asked Questions about feature-spec

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I write a feature specification before implementation?▼

Start by exploring the codebase to understand existing patterns, then clarify scope through focused questions, propose 2-3 approaches with trade-offs, and align section by section with stakeholders. The final spec should contain a problem statement, numbered requirements, scope boundaries, and verifiable acceptance criteria.

What should a good feature specification document include?▼

A feature specification should include a problem statement, numbered testable requirements, explicit in-scope and out-of-scope lists, acceptance criteria tied to each requirement, and constraints. Open questions must be resolved before implementation planning begins.

How do I prevent scope creep in feature requirements?▼

Apply YAGNI strictly: only specify what the user explicitly needs right now and cut nice-to-haves or future-proofing. Assess every requirement against the confirmed scope before including it, and explicitly list exclusions in an Out of Scope section.

Should a feature spec include technical implementation details?▼

No. Specifications describe WHAT and WHY from the user's perspective, never HOW. Implementation details, technical decisions, and code structure belong in the implementation plan, which is a separate downstream document.

When is a feature specification not ready for implementation planning?▼

A spec is not ready when it contains unresolved open questions, unverifiable requirements, or ambiguous scope boundaries. Every requirement must be testable, and the Open Questions section must be empty or removed before planning begins.