design

Author codebase-grounded feature design documents in .design/<slug>.md.

27|3|Updated Mar 5, 2026
One-click install
npx skills add https://github.com/forecast-bio/ferray --skill design-forecast-bio
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: design
Source: https://github.com/forecast-bio/ferray/tree/main/.claude/skills/design
Command: npx skills add https://github.com/forecast-bio/ferray --skill design-forecast-bio

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Use to author or iterate a feature design document grounded in actual codebase exploration. Drives a four-phase flow (Explore & Interview → Draft → Resolve open questions → Iterate) and writes .design/<slug>.md plus a pipeline state file ready for crosslink kickoff --doc. Trigger when the user says "design <feature>", "/design …", "iterate the design", or asks to write a design doc / spec / RFC for a feature.

Core Features & Use Cases

  • Structured, codebase-grounded design documents for new features.
  • Four-phase design process: explore/interview, draft, resolve open questions, iterate.
  • Generates a .design/<slug>.md node and a pipeline state file ready for kickoff.

Quick Start

Draft a design document for a chosen feature by following the four-phase flow and store it in .design/<slug>.md.

Frequently Asked Questions about design

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

FAQPage Schema
How do I write a feature design document grounded in actual codebase exploration?

Author a codebase-grounded feature design document by following a four-phase flow: explore and interview, draft, resolve open questions, and iterate. This process ensures your design defines concrete requirements, architecture references, and acceptance criteria.

What is the best way to iterate on an existing design doc or RFC for a new feature?

Iterate on a design doc by triggering the iteration flow, which revisits open questions and updates the design file. This refines the existing architecture references and requirements to produce an updated, kickoff-ready pipeline state.

How does codebase exploration work when drafting a design spec?

Codebase exploration works by grounding the design interview and drafting phases in actual project code. This mechanism ensures the generated design document references existing architecture and accurately defines concrete feature requirements.

Can I use this design process to generate a pipeline state ready for kickoff?

Yes, the design process generates a .design/<slug>.md file and a pipeline state file specifically ready for kickoff. This output ensures your feature design transitions directly into downstream development workflows.

Do I need to resolve all open questions before finalizing a feature design document?

No, the design process explicitly captures unresolved items as open questions to resolve during iteration. This approach lets you finalize the initial draft while tracking specific architectural or requirement uncertainties for later.

What are the limitations of authoring design docs without codebase exploration?

Authoring design docs without codebase exploration risks missing concrete architecture references and accurate acceptance criteria. Grounding the document in actual code ensures the feature design defines feasible requirements and resolves true open questions.