heliohq/ship / skills/design
design
VerifiedPlan implementation before coding: investigate the repo, write spec and plan, and validate with a peer. Use for "plan", "design approach", "scope", or any coding task needing a plan. Not system-design thinking (/ship:arch-design) or full /ship:auto.
| File | Kind | Size |
|---|---|---|
| SKILL.md | skill md | 15.9 KB |
| execution-drill.md | markdown | 3.7 KB |
| independent-investigator.md | markdown | 5.2 KB |
| write-plan.md | markdown | 6.0 KB |
| write-spec.md | markdown | 1.6 KB |
write-spec.md
View on GitHub# Write Spec — Investigation & Spec Authoring
How to investigate a codebase and write a spec. Used by the host agent
in Phase 2-3 of `/ship:design`.
## Overview
Write the spec assuming the reader has zero context: what you found,
what you traced, and what must be true for the task to be done.
## Methodology
Your investigation method and spec structure are the **same** ones you
dispatch to the peer — the steps are role-neutral. Follow the
**Investigation** and **Write Spec** sections of the peer prompt in
`independent-investigator.md` (the same-session file you read to dispatch
the peer): trace callers backward and consumers forward, search for
existing defenses before proposing a fix, verify a file exists before
proposing to create it, grep tests that assert values you'll change,
cross-reference every consumer of a changed interface — then the "what to
include" section list and the spec self-review checklist.
Ignore the peer-framing ("you have not seen any prior spec"); apply the
steps to your own investigation and write the result to `spec.md`.
**This is the most important phase — do not rush the investigation.**
Every claim in the spec must reference a `file:line` you actually read.
## Task too vague?
After investigation, check if any of these are missing from the task
description AND could not be inferred from code:
- **Target behavior** — what should change
- **Target surface** — which files, endpoints, or components
- **Success condition** — how to know it's done
If any are missing, ask the user via AskUserQuestion before writing the
spec. (This is host-only — the peer investigator cannot ask the user.)
Related skills
| Skill | Stars | Updated | Works with | Status | Install |
|---|---|---|---|---|---|
| brainstormingobra/superpowersYou MUST use this before any creative work - creating features, building components, adding functionality, modifying behavior, or planning anything new, in software or out of it (a talk, a business, a renovation). | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |
| brand-guidelinesanthropics/skillsApplies Anthropic's official brand colors and typography to any sort of artifact that may benefit from having Anthropic's look-and-feel. Use it when brand colors or style guidelines, visual formatting, or company design standards apply. | Stars: 198k | Updated | Claude CodeCodex | Verified | Install |
| frontend-designanthropics/skillsGuidance for distinctive, intentional visual design when building new UI or reshaping an existing one. Helps with aesthetic direction, typography, and making choices that don't read as templated defaults. | Stars: 198k | Updated | Claude CodeCodex | Verified | Install |
| mcp-builderanthropics/skillsGuide for creating high-quality MCP (Model Context Protocol) servers that enable LLMs to interact with external services through well-designed tools. Use when building MCP servers to integrate external APIs or services, whether in Python (FastMCP) or Node/TypeScript (MCP SDK). | Stars: 198k | Updated | Claude CodeCodex | Verified | Install |