By korflux
Keel — five commands covering the software development lifecycle: spec, plan, build, test, review.
Implement tasks incrementally — build, test, verify, commit. Add "auto" to run the whole plan in one approved pass.
Extract what the user actually wants, one question at a time, then refine a vague direction into a concrete one
Break work into small verifiable tasks with acceptance criteria and dependency ordering
Conduct a five-axis code review — correctness, readability, architecture, security, performance
Simplify code for clarity and maintainability — reduce complexity without changing behavior
Manages deprecation and migration. Use when removing old systems, APIs, or features. Use when migrating users from one implementation to another. Use when deciding whether to maintain or sunset existing code.
Guides stable API and interface design. Use when designing APIs, module boundaries, or any public interface. Use when creating REST or GraphQL endpoints, defining type contracts between modules, or establishing boundaries between frontend and backend.
Tests in real browsers via Chrome DevTools MCP. Use when building or debugging anything that runs in a browser. Use when you need to inspect the DOM, capture console errors, analyze network requests, profile performance, or verify visual output with real runtime data. Requires the chrome-devtools MCP server to be configured.
Automates CI/CD pipeline setup. Use when setting up or modifying build and deployment pipelines. Use when you need to automate quality gates, configure test runners in CI, or establish deployment strategies.
Optimizes agent context setup. Use when starting a new session, when agent output quality degrades, when switching between tasks, or when you need to configure rules files and context for a project.
Based on adoption, maintenance, documentation, and repository signals. Not a security audit or endorsement.
A seven-command lifecycle that carries a change from raw idea to merge, for AI coding agents.
Every skill's content — the workflows, checklists, and engineering judgment — was written by Addy Osmani and the original contributors. This fork doesn't change what the skills teach. It reorganizes how you reach them: 24 loosely-connected skills become a six-phase lifecycle (keel-interview → keel-spec → keel-plan → keel-build → keel-test → keel-review), plus keel-simplify as a run-anytime utility, with the specialist skills wired in as on-demand call-ins instead of a flat list you have to memorize. Same discipline, easier to actually use.
This is a fork. Keel (repo:
korflux/keel-skills) is a personal adjustment of the original Addy Osmani / agent-skills project, maintained by @korflux. All skill design and workflow content are the work of Addy Osmani and the original contributors (see Credits) — this fork renames commands so the same short name works the same way in Claude Code, Grok Build, Antigravity, and Codex, drops the persona//shiplayer in favor of on-demand skills wired directly into the lifecycle commands, and organizes the catalog around the seven-command flow above. For the canonical, full-featured upstream project, go to addyosmani/agent-skills.
INTERVIEW DEFINE PLAN BUILD TEST REVIEW
┌────────┐ ┌──────┐ ┌──────┐ ┌──────┐ ┌──────┐ ┌──────┐
│ Extract│───▶│ Spec │ ───▶ │ Plan │ ───▶ │ Code │ ───▶ │ Test │ ───▶ │ QA │
│ intent │ │ PRD │ │ Tasks│ │ Impl │ │Debug │ │ Gate │
└────────┘ └──────┘ └──────┘ └──────┘ └──────┘ └──────┘
/keel-interview /keel-spec /keel-plan /keel-build /keel-test /keel-review
7 slash commands: the 6-phase lifecycle, plus /keel-simplify as a standalone utility you can run any time.
| What you're doing | Command | Key principle |
|---|---|---|
| Find out what's actually wanted | /keel-interview | Ask before you assume |
| Define what to build | /keel-spec | Spec before code |
| Plan how to build it | /keel-plan | Small, atomic tasks |
| Build incrementally | /keel-build | One slice at a time |
| Prove it works | /keel-test | Tests are proof |
| Review before merge | /keel-review | Improve code health |
| Simplify the code | /keel-simplify | Clarity over cleverness |
Want fewer manual steps once the spec exists? /keel-build auto generates the plan and implements every task in a single approved pass — you approve the plan once, then it runs autonomously. It removes the human stepping between tasks, not the verification: every task is still test-driven and committed individually, and it pauses on failures or risky steps.
Every command file already carries the keel- prefix, so the same short name works whether your CLI resolves it as a slash command or as a $skill reference — no per-tool renaming needed to dodge collisions.
Once a command is loaded (see Install below), the invocation looks almost identical across tools because the command name itself carries the brand prefix:
| Command | Claude Code | Grok Build | Antigravity | Codex |
|---|---|---|---|---|
| Interview | /keel-interview | /keel-interview | /keel-interview | $keel-interview |
| Spec | /keel-spec | /keel-spec | /keel-spec | $keel-spec |
| Plan | /keel-plan | /keel-plan | /keel-plan | $keel-plan |
| Build | /keel-build | /keel-build | /keel-build | $keel-build |
| Test | /keel-test | /keel-test | /keel-test | $keel-test |
| Review | /keel-review | /keel-review | /keel-review | $keel-review |
| Simplify | /keel-simplify | /keel-simplify | /keel-simplify | $keel-simplify |
Codex has no slash-command concept at all — its plugin manifest has no commands field, so a skill is invoked with $name instead. Claude Code and Grok Build support a qualified keel:keel-spec form if you ever need to disambiguate against another installed plugin, but day to day the unqualified /keel-spec is what you'll type.
Recommended — verified, works with 70+ agents, no plugin registration needed. The skills CLI clones this repo and drops the skills you pick into whichever agent(s) it detects on your machine (Claude Code, Cursor, Codex, Copilot, Cline, and more):
npx claudepluginhub korflux/keel-skills --plugin keelLazy senior dev mode. Forces the simplest, shortest solution that actually works: YAGNI, stdlib first, no unrequested abstractions.
Comprehensive skill pack with 66 specialized skills for full-stack developers: 12 language experts (Python, TypeScript, Go, Rust, C++, Swift, Kotlin, C#, PHP, Java, SQL, JavaScript), 10 backend frameworks, 6 frontend/mobile, plus infrastructure, DevOps, security, and testing. Features progressive disclosure architecture for 50% faster loading.
Evidence-gated AI coding workflow: scan → analyze → plan → TDD → execute → fix → verify → review, powered by Codebase Memory MCP >= 0.9.0 with optional Serena LSP intelligence. Includes blast-radius planning, test/cycle gates, independent review, and Windows Git Bash hook auto-resolution.
Consult multiple AI coding agents (Gemini, OpenAI, Grok, Perplexity, plus codex, antigravity, and grok CLIs when installed) to get diverse perspectives on coding problems
Production-grade engineering skills for AI coding agents — covering the full software development lifecycle from spec to ship.
Feature development with code-architect/explorer/reviewer agents, CLAUDE.md audit and session learnings, and Agent Skills creation with eval benchmarking from Anthropic.