Ontology Spec

Write and maintain an ontology-based design-doc set (ONTOLOGY.md + CRITERIA.md + RATIONALE.md + ARCHITECTURE_LOG.md, plus DYNAMICS.md for games) instead of a single SPEC.md. ONTOLOGY is the implementation reference — purpose, entities as structs, invariants, rules as short pseudocode, commands and events, numbers — with provenance tags ([D]/[P]/[T]) and Serves tags; CRITERIA turns every rule into input → expect rows, each owning a test directory; RATIONALE records the why of current decisions with alternatives and evidence; ARCHITECTURE_LOG holds the dated history. Use when starting a new app,

Install

skills CLI (Claude Code, Cursor, Codex and more)
npx skills add roby2358/skills --skill ontology_spec
Manual install
git clone --depth 1 https://github.com/roby2358/skills.git /tmp/skills
mkdir -p ~/.claude/skills
cp -r /tmp/skills/ontology_spec ~/.claude/skills/ontology_spec

Personal skills live in ~/.claude/skills. Use .claude/skills in a project folder to share a skill with your team.

From the SKILL.md

# Ontology Spec

A design-doc set that separates **what exists and how it behaves** (ONTOLOGY), **how we
know it's right** (CRITERIA), **why each choice was made** (RATIONALE), and **how it got
here** (ARCHITECTURE_LOG). For games and other systems with emergent behavior, add **why
it's fun / why it works** (DYNAMICS).

The gains come from four properties, not from the number of files:
1. **Tagged provenance** — the designer's decisions are never confused with Claude's guesses.
2. **Rationale with alternatives** — writing down what else was considered finds bugs.
3. **Criteria as tables** — rules arrive with their test cases, so tests are transcription.
4. **Stable IDs** — code, tests and docs cite `G-n`/`D-n`/`Q-n`/`R-n`, which never move.

If a project is a small CRUD feature, use the `specification` skill (SPEC.md) instead, and
borrow these properties as sections. Use this format when the *why* matters as much
as the *what*: games, simulations, rule engines, protocols, security services, anything
that will be tuned.

After the docs exist, build with `/implement <app>`; it reads this format.

---

## The documents

| Doc | Answers | Reader | Pace |
|---|---|---|---|
| **ONTOLOGY.md** | What exists, what each thing carries, how things relate, the rules, the interface | Implementer | The real spec. Consulted constantly. |
| **CRITERIA.md** | How we know it's correct, readable, and good | Tester / tuner | Written with the rules; each row owns its tests |
Read the full skill on GitHub

Details

Skill name
ontology_spec
Path
ontology_spec/SKILL.md
Repo updated
14 November 2026

FAQ

What does the Ontology Spec skill do?

Write and maintain an ontology-based design-doc set (ONTOLOGY.md + CRITERIA.md + RATIONALE.md + ARCHITECTURE_LOG.md, plus DYNAMICS.md for games) instead of a single SPEC.md. ONTOLOGY is the implementation reference — purpose, entities as structs, invariants, rules as short pseudocode, commands and events, numbers — with provenance tags ([D]/[P]/[T]) and Serves tags; CRITERIA turns every rule into input → expect rows, each owning a test directory; RATIONALE records the why of current decisions with alternatives and evidence; ARCHITECTURE_LOG holds the dated history. Use when starting a new app,

How do I install the Ontology Spec skill?

skills CLI (Claude Code, Cursor, Codex and more): npx skills add roby2358/skills --skill ontology_spec. Skills go in ~/.claude/skills (personal) or .claude/skills (project), and Claude loads them automatically when a task matches the description.

Is Ontology Spec popular?

It lives in roby2358/skills, which has 5 GitHub stars and 0 forks.

Related skills

Coding

5

roby2358/skills

Project-specific coding conventions and constraints. Use when writing, modifying, or refactoring any code, implementing features, writing tests, or building MCP servers in this project.

Software Development

Game Design

5

roby2358/skills

Design principles for tactics and strategy games. Use when designing new mechanics, tuning existing systems, adding units or abilities, adjusting difficulty, or making any game design decision. Triggers on discussions about game balance, unit roles, terrain, pacing, spawning, healing, combat, win/loss conditions, or UI for conveying game state.

Frontend & Design

Game Productize

5

roby2358/skills

Audit a browser game for ship-readiness and produce a tiered gap list: a SHIP.md in the game repo plus pjpd tasks. Finds what gets forgotten between "the game works" and "a stranger can play it": title/menus/end screens, save and resume, audio and settings, onboarding, the mobile "sorry" screen, localization readiness, error recovery, hosting on the author's site and on portals (itch.io, Newgrounds), credits, asset licenses, and AI disclosure. Use when the user says "productize", "/game-productize", "get this game ready to ship/release/publish/launch", "what's missing before I launch", "prep f

Documents & Office Files

Openapi Spec Generation

9

sandbaseai/workbuddy-skill

Use when creating, generating, reviewing, or maintaining an OpenAPI 3.1 contract from code or design, including API documentation, SDK generation, and implementation validation.

Software Development

Oauth Debugging

9

sandbaseai/workbuddy-skill

Use when an OAuth 2.0 or OpenID Connect flow fails, loops, loses session state, rejects a token, or grants the wrong access; reconstruct the exact protocol exchange without exposing credentials or live tokens.

Software Development

Openapi to Application Code

9

sandbaseai/workbuddy-skill

Use when turning an OpenAPI specification into an application scaffold or reviewing generated handlers, models, tests, and configuration.

Software Development

Browse the most popular Claude skills, all software development skills, or write your own with the skills generator.