Summary
This is the entry point for Ralph Specum's spec workflow. It parses your goal, sets up the spec directory structure, initializes state tracking in `.ralph-state.json`, and writes a `.progress.md` file. By default it stops after setup and waits for you to explicitly continue, which is actually helpful because you can review the generated artifacts before it starts tearing through your codebase. Quick mode (`--quick`) skips that pause and runs straight into implementation. The state tracking is pretty thorough: phase, task index, iteration counts, commit flags, and related specs. It also handles epic context and will route large goals to triage instead of cramming everything into one spec. Honest take: the explicit approval gates feel bureaucratic until you've seen an LLM misinterpret a vague goal.
Install to Claude Code
npx -y skills add tzachbon/smart-ralph --skill ralph-specum-start --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 →
