295+ skills indexed
A curated catalog of drop-in capabilities across business, design, engineering, marketing, and ops — searchable and growing.
Product · Free
Terminal Skills
Open registry of drop-in capabilities for Claude Code, Codex, Gemini CLI, and Cursor. Install a specialised skill with one command.
Terminal Skills is a public registry of pre-built skills that extend coding agents with specialised, repeatable workflows — from A/B test setup and analytics tracking to AI-generated slide decks. Browse, install, and your agent picks up the new capability the next time it runs.
The premise: agents don't need to be smarter, they need to be equipped. Every skill is open source under Apache 2.0, readable by the agent directly, and installable with a single command — no accounts, no paywalls, no lock-in.
What it does · 02
Built for the job, end to end.
Install in seconds
Paste a single command — `terminal-skills install <skill>` — or use the meta-skill for guided setup. No accounts to create.
Works across coding agents
Skills run in Claude Code, Codex, Gemini CLI, and Cursor. Author once, ship everywhere your team works.
Apache 2.0, readable by the agent
Every skill ships as readable source. Audit before you install, fork to customise, contribute back when you improve it.
Workflows, not just prompts
A skill bundles instructions, examples, and tool patterns for one job — A/B tests, schema markup, paid ads, onboarding flows — so the agent ships like a senior.
No accounts, no paywalls
The registry, the CLI, and every skill are free. You own what you install — and what you build on top.
Inside · 03
A look at the workspace.
Fig. 01/04
Skills for the specialised agent — drop-in capabilities for Claude Code, Codex, Gemini CLI, and Cursor.
Fig. 02/04
Browse the registry — search across 247+ skills, filter by category (Business, Marketing, Design, Operations…).
Terminal Skills registry browse page with search, category filters, and top-performing skills
Fig. 03/04
Every skill ships with trust signals — quality, security, and measured agent-success uplift.
Skill detail page for ab-test-setup showing trust score, quality, security, and impact metrics
Fig. 04/04
One command. Whole workflow. Install via the meta-skill or the CLI — the agent does the rest.
Install section showing one-command install via meta-skill and CLI patterns