claudekit

Shared skills, commands, and agents for AI coding harnesses. Synced across machines via iCloud.

Setup

All directories are symlinked from each harness's config into this repo:

~/.claude/skills/    → claudekit/skills/     (Claude Code)
~/.claude/commands/  → claudekit/commands/    (Claude Code)
~/.claude/agents/    → claudekit/agents/      (Claude Code)
~/.codex/skills/     → claudekit/skills/      (Codex)
~/.config/opencode/skills/  → claudekit/skills/   (OpenCode)
~/.config/opencode/agent/   → claudekit/agents/   (OpenCode)
~/.config/opencode/command/ → claudekit/commands/  (OpenCode)

Codex only supports skills. Claude Code and OpenCode support all three.

Skills (21)

Auto-triggered background knowledge loaded when the agent detects a matching task.

Skill	Description
ashby	Ashby ATS API reference and recruiting workflows
atlas	macOS AppleScript control for ChatGPT Atlas app
cloudflare	Workers, Pages, Sandbox SDK, MCP client, full platform
council	Multi-agent codebase exploration (spawns n=10 agents)
docs	Write, update, or generate state diagrams for docs
figma	Figma-to-code via Figma MCP server
frontend-design	Design philosophy, constraints, accessibility, React/Next.js
github	Fix CI, address PR comments via gh CLI
himalaya	CLI email client (IMAP/SMTP/Maildir)
humanizer	Remove AI-writing patterns from text
imagegen	Generate/edit images via OpenAI Image API
investment	Populate investment memos + add to portfolio database
missive	Missive inbox management, drafting, labeling, contacts
openai-docs	Query OpenAI developer docs MCP
pdf	Read, create, fill, sign PDF/Word documents
playwright	Browser automation via playwright-cli
shaping	Collaborative solution shaping, breadboarding, slicing
spreadsheet	Create/edit/analyze xlsx/csv with Python
transcribe	Audio-to-text with diarization
vercel-deploy	Deploy to Vercel via bundled deploy script
yeet	Stage, commit, push, open PR in one flow

Commands (11)

User-invoked /command actions.

Command	Description
/ashby	Search candidates, view jobs, manage pipeline
/audit	3x parallel code reviewers on changes or PRs
/extract-pattern	Find implementation patterns across your GitHub repos
/fal	Generate images, videos, audio, 3D via fal.ai
/fix	Auto-fix issues found by /audit
/frontend-design	Set up frontend project (shadcn, Tailwind, etc.)
/ios-release	Build and upload iOS app to App Store Connect
/missive	Review inbox, draft replies, or show summary
/restaurant-research	Multi-platform restaurant research with pricing
/team	Create an agent team for a task
/vercel-debug	Diagnose and fix Vercel build failures

Agents (4)

Specialized personas spawned as subagents with specific tools and models.

Agent	Model	Description
engineer	inherit	Implementation, fixes, debugging (4 modes)
frontend-engineer	opus	Distinctive UI with browser tools
code-reviewer	inherit	Code review and codebase audits
auth-expert	opus	Auth debugging (PKCE, OAuth, cookies, sessions)

Adding Components

Skill

Create skills/<name>/SKILL.md:

---
name: skill-name
description: Triggers for this skill...
---

Instructions here...

Command

Create commands/<name>.md:

---
description: What the command does
argument-hint: [optional-arg]
allowed-tools: Tool1, Tool2
---

Instructions here...

Agent

Create agents/<name>.md:

---
name: agent-name
description: When to use this agent...
tools: Read, Write, Edit, Bash
model: inherit
color: red
skills:
  - skill-name
---

Instructions here...

Name	missive-inbox-triage
Description	This skill should be used when the user asks to "triage my inbox", "prioritize emails", "what needs my attention", "sort through my messages", "help me process my inbox", or is reviewing conversations to determine what needs action in Missive.

missive-inbox-triage

SKILL.md