Skip to content
adeptabilityadeptability

adeptability

Write an AI coding skill once, run it in every AI coding assistant.

adept is a small Go CLI that lets you author an AI agent skill — a prompt, rule, or procedure — once, in a single canonical format, then render it accurately into Claude Code, Cursor, GitHub Copilot, OpenAI Codex, OpenCode, and 50+ other AI coding agents.

Every agent loads skills from a different path with a different schema (.claude/skills/<id>/SKILL.md, .cursor/rules/<id>.mdc, a single AGENTS.md for Codex, .github/instructions/*.instructions.md for Copilot, …). Keeping the same knowledge consistent across all of them by hand drifts fast. adept keeps one source of truth and handles the per-harness frontmatter, activation rules, aggregation, and size budgets.

Think of it as dotfiles for your AI coding agents.

Why adept

Section titled “Why adept”
  • One skill, every format. No copy-pasting incompatible frontmatter into five paths.
  • Drift-proof. Edit a skill in any harness, pull it back to canonical, re-publish. A hash-based 3-way detector flags divergence via adept status and adept diff.
  • Package-manager libraries. Share a versioned set of skills across a team by reference — the repo commits a pointer, not the bytes.
  • Safe installs. A static + optional-LLM scanner checks skills from skills.sh / GitHub before they land; critical findings hard-block by default.
  • Signed & reproducible. Content-hashed skills, pinned upstream provenance, cosign-signed release binaries.

Where to next

Section titled “Where to next”
  • 📦 Install — Go, Homebrew, curl, Docker.
  • 🚀 Quickstart — author, sync, and see it in your harness in 60 seconds.
  • 💡 Concepts — canonical skills, harnesses, libraries, layouts, drift.
  • ⌨️ Command reference — every command and flag.

The core loop

Section titled “The core loop”
Terminal window
adept init                          # scaffold, adopt existing harness files, seed default skills
adept skill add lint-style --edit   # author a canonical skill
adept harness add claude-code
adept harness add cursor
adept sync                          # render canonical → every enabled harness
adept status                        # init, libraries, harnesses, and drift at a glance

Author in one place, adept sync, and each agent gets its own correct format.