Home/Architecture
Repository architecture

One repo, every agent

claude-skills is a Claude Code plugin marketplace, a home for standalone tools (the Second Wind CLI and the Cache Clock extension for Copilot CLI), and a set of adapters that wire the same skills into Codex, Copilot CLI, and any AGENTS.md-aware agent — plus private seed machinery that rebuilds a full dev environment on a fresh machine.

System overview

flowchart TD
  MP["marketplace.json
.claude-plugin/"]
  IOS["ios-dev
59 skills · 6 cmds"]
  LIN["linear-pm
1 skill · 6 cmds"]
  CW["core-workflow
19 skills · 4 cmds · 2 agents · 1 hook"]
  SW["second-wind
skill wrapper"]
  PC["prompt-craft
7 skills · 6 hooks"]
  PA["portable-artifacts
5 skills · installer"]
  TB["test-board
test-run band"]
  SC["session-compass
compass band · /compass pane"]
  WIND[("tools/second-wind
wind.py")]
  CCLK[("tools/cache-clock
Copilot CLI extension")]
  AD["adapters/install.sh"]
  SEED["setup/
setup.sh · capture.sh · contribute.sh"]
  CC(("Claude Code"))
  OTHER(("Codex / Copilot /
AGENTS.md agents"))

  MP --> IOS & LIN & CW & SW & PC & PA & TB & SC
  SW -.->|"wraps"| WIND
  CC -->|"/plugin marketplace add"| MP
  SEED -->|"local_plugins step"| CC
  SEED -->|"wind → ~/.local/bin"| WIND
  AD -->|"symlink skills"| OTHER
  CCLK -->|"install.sh → ~/.copilot/extensions"| OTHER
  IOS & LIN & CW & SW & PC & PA -->|"skills/"| AD

  classDef plugin fill:#3b6a7a1a,stroke:#3b6a7a,stroke-width:2px
  classDef infra fill:#8f54101a,stroke:#8f5410,stroke-width:2px
  class IOS,LIN,CW,SW,PC,PA,TB,SC plugin
  class SEED,AD infra

The eight plugins

ios-dev

build → screenshot → phone → ship

  • app-preview — build on simulator, screenshot, deliver to iPhone (iMessage ping + iCloud Drive)
  • ios-build — sim/device builds with signing auto-detection
  • release — archive → .ipa → validate → upload to App Store Connect
  • /preview, /fix commands — screenshot to your iPhone in one shot

linear-pm

issue conventions + autonomous pickup

  • linear-pm skill — label vocabulary, status taxonomy, branch/PR naming, .claude/linear.yml policy
  • 6 commands — init, new, pick (the autonomous loop), status, sync, block

core-workflow

everyday glue

  • commit — shellcheck-gated conventional commits
  • contribute — push changes back to this repo from anywhere
  • agents — image-parser (vision), web-researcher (docs lookup)
  • hook — shellcheck on every .sh Edit/Write

second-wind

outlast the 5-hour limit

  • wind CLI — one tmux session per repo, watches for the usage-limit pause, resumes everything after reset
  • skill wrapper — teaches the agent wind init/add/up/watch/status/resume/down + self-install fallback

prompt-craft

rough ask → deterministic spec

  • /improve-prompt — restated goal, acceptance criteria, assumptions, recommended commands
  • /plan — decompose into per-step acceptance criteria + a TodoWrite plan
  • debug / refactor / review lenses — correctness + security first
  • /refresh — re-scan skills and rebuild the command-advisor registry + history profile on demand; also wires the statusline
  • command advisor — ranks the most relevant next command by combining keyword relevance, your command-history frequency, and git-state context (at most 3 recommendations). It powers the pre-prompt banner, the Stop-hook follow-ups, and the statusline segment. Every surface is user-only — advisor output never enters the model context.
  • hooks — suggest_next (Stop, now routed through the command advisor), prompt_hint (UserPromptSubmit), registry_freshness (SessionStart), statusline_hint (statusLine) + opt-in block_secrets / format_on_edit

test-board

pass/fail band after every test run

  • band — passed/failed/skipped counts above the prompt, next to the previous run of the same suite in the same worktree

session-compass

what each session is building and where it stands

  • compass band, /compass pane and compass_status tool — what each session is building, where it stands and what it produced

portable-artifacts

dashboards + diagrams + explainer pages + 3D scenes + animated videos, offline

  • /make-dashboard — prompt + data file → SQL + Vega-Lite spec → one static HTML page; every chart shows its SQL and refresh time
  • /make-diagram — prompt → JSON topology → ELK-laid-out HTML diagram; zoom, a story player, an Explore panel, full screen
  • /make-explainer — prompt → typed JSON sections → one HTML page or slide deck: branch recap, concept explainer, plan, test checklist with saved progress; embeds diagrams, charts, 3D scenes and video
  • /make-motion — prompt → scene spec → live preview + deterministic 1080p MP4; silent unless you ask for sound, then music from your file or a voiceover synthesised locally with Kokoro
  • /make-scene — prompt → JSON spec → one offline three.js page: a city, forest, galaxy or terrain from your data, or a diagram or dashboard chart lifted into 3D, with a camera tour; the build offers the other worlds
  • bin/install.mjs — copies all five skills into Claude Code, Codex, Copilot or Gemini CLI skill directories

How installs flow

Public user — Claude Code

anyone, two commands

/plugin marketplace add abhijitbansal/claude-skills → /plugin install <name>@claude-skills → skills + commands + hooks live
/plugin marketplace add abhijitbansal/claude-skills
/plugin install core-workflow@claude-skills

Public user — other agents

clone, then one script

git clone → adapters/install.sh codex | copilot | agents-md → same SKILL.md files, symlinked or indexed

Owner — fresh machine seed

setup.sh, eight idempotent steps

preflight→ claude→ marketplaces→ plugins→ skills→ dotfiles→ local_plugins→ symlinks
One source of truthThe owner's machines install the repo's own plugins through the same marketplace mechanism the public uses (local_plugins step) — there is no second code path to drift.

Markdown twin: docs/architecture.md · Catalog: skills & tools catalog · Repo: github.com/abhijitbansal/claude-skills