Summary
This is a complete design system for building CLIs that agents can actually use. Every command returns structured JSON with HATEOAS-style next_actions arrays that tell the agent what to run next, using POSIX templates with pre-filled context values. No plain text, no tables, no --json flags. The envelope format is strict (ok/command/result/next_actions for success, plus error/fix for failures) and backed by Effect CLI with Bun. Output is context-protecting by default, auto-truncating logs and large responses to avoid blowing up token budgets. Use this when building new CLIs or adding commands to existing ones like joelclaw or slog. The opinionated constraints make it narrow but extremely practical for agentic workflows.
Install to Claude Code
npx -y skills add joelhooks/joelclaw --skill cli-design --agent claude-codeInstalls into .claude/skills of the current project.
Keep your Mac awake while Claude Code and 40+ AI agents run. Sleeps when they're idle.
One time payment $9 →
Integrate web data into your AI product. One API to scrape website & brand data.
Get API Key Now →
Agent, run crypto. Access onchain data & trade routes via 1inch.
Install now →
On Capafy, your Skill runs online 24/7 as an agent product, and you get paid every time someone uses it.
Start earning →
Files
SKILL.md
Select a file.
Featured
Keep your Mac awake while Claude Code and 40+ AI agents run. Sleeps when they're idle.
One time payment $9 →Integrate web data into your AI product. One API to scrape website & brand data.
Get API Key Now →Agent, run crypto. Access onchain data & trade routes via 1inch.
Install now →On Capafy, your Skill runs online 24/7 as an agent product, and you get paid every time someone uses it.
Start earning →
