Guizang's agent skill turns your repo's recent updates into a finished promo video — real components, code-rendered animation, an original code-written soundtrack — no video editor involved.
- agent-skill
- video
- promo
- marketing
An ontology-driven development skill: requirements exploration → seven-model ontology YAML → a runnable management system, with a human sign-off gate at every stage.
ontology-driven-dev turns a business requirement into a runnable management system where every feature traces back to a requirement item. Seven ontology models (M1/M2/M3/M5/M6/M7/MU) in YAML are the single semantic source: database tables, APIs, menus, permissions, workflows, and rules all generate from the models, so the model and the code cannot drift apart. Stage one's eight exploration phases each pause for human confirmation, and the generated app ships with a built-in AI chat panel backed by a read-only SQL boundary.
For internal business systems where requirement traceability and model-code alignment matter more than framework choice: approval flows, contract management, back-office tools — the class of app where 'the code drifted from the spec' is the recurring failure.
# Claude Code (user level) cp -r ontology-driven-dev ~/.claude/skills/ontology-driven-dev # or project level cp -r ontology-driven-dev .claude/skills/ontology-driven-dev # Codex: copy into the repo and reference SKILL.md from codex.md / AGENTS.md
Documentation is Chinese-first (an English README is included). The stack is fixed to its code-paas base rather than your own framework, and generated apps need network access for npm/pip installs — Codex sandboxes must allow it. The human gates are the point: budget real review time per stage.
Repository
sharptoolbox/ontology-driven-devCharted
Guizang's agent skill turns your repo's recent updates into a finished promo video — real components, code-rendered animation, an original code-written soundtrack — no video editor involved.
An agent skill that geolocates a photo with nothing readable in it — OSM geometry, elevation skylines and satellite tiles turn “where was this taken?” into a measured answer with an error radius.
A skill that makes AI coding agents map the architecture and declare the blast radius before they touch a line of code — rendered as one standalone HTML page.