Productivity

AI 开发完整流程：规格驱动(SDD)+分层测试先行(TDD)+收敛验证，让 AI 生成规范不跑偏的代码。含 9 份中间产物模板与可安装 Agent Skill。

A safety-first Codex skill for Git, GitHub/Gitea, HANDOFF.md, and two-computer project handoffs on Windows.

Generate the development environment for AI-assisted development — docs, Cursor rules, Claude skills, prompts and checklists tailored to your stack.

A Claude Skill that turns any job posting into a structured Notion record: normalized fields, extracted and categorized skills, a gap check against your own stack, and a cumulative skills bank that ranks what the market actually asks for. Fully customizable fields and ranking criteria.

Codex skill for a rigorously sourced weekly Top 15 of physical Maker Projects

Clone-and-go project scaffolder: setup.ps1 + justfile + .docs + Claude skills for 9 stacks

Windows Codex Skill for safely pausing, checking in, and checking out complete project folders across PCs.

Custom Claude Code skills for analytical and engineering workflows

Reusable personal Codex workflow skills

A Codex skill for deterministic project documentation governance, freshness checks, and archive isolation.

A workshop where skills for LLM agents are engineered, not just written: developed test-first against baseline agent behavior, hardened by adversarial review, and required to earn their keep with a logged field-win record.

A Codex skill for evaluating and migrating legacy Android projects to Kotlin Multiplatform

Portable, verified AI skills from the DOL workflow

Reusable AI skills for prompt refinement based on real-world AI/ML project challenges.

My Claude Skills for Productivity

Agent Skills for Scenario: game asset, texture, skybox, 3D and model-training workflows over the Scenario MCP. Install: npx skills add scenario-labs/skills

A Codex skill for human-led AI evolution through task review, memory candidates, external evaluation, and safe skill upgrades.

Two Claude Code Skills for Codex collaboration: execution delegation discipline + plan co-creation discipline, avoiding context pollution and false second opinions in dual-model workflows

JRM Studio — shared identity, community health files, canonical AI layer (agents/skills/prompts), reusable workflows, and the cross-repo sync tool.

Reusable Codex skills for Claire's workflows.

Private Codex skills for proxy and VPS configuration workflows

Codex skill: send a detailed QQ email notification after a goal/task completes.

Personal OpenAI/Codex skills for reusable workflows

[AI agents] Installable GitHub library of 1,400+ agentic skills for Claude Code, Cursor, Codex CLI, Gemini CLI, Antigravity, and more. Includes installer CLI, bundles, workflows, and official/community skill collections.