Shipping claude-shiploop
Every case study on this site runs through the same loop: classify the request, write the spec, build to the spec, measure the result, promote what repeats, forget what doesn't. That loop had a working name — ShipLoop — a one-page protocol in CLAUDE.md , five skills in .claude/skills/ , and a scorecard that reads commits, DEVLOG headings, and session logs to measure how often each project…
The <source> block details a protocol called ShipLoop, which helps users classify requests, write specifications, build projects to specifications, measure results, and promote what works while discarding what doesn't. The protocol consists of three components: the protocol itself, five skill commands, and a scorecard. The protocol is a nine-lane request classifier and a spec-plan-tasks gate.
The five skills are entry-point slash commands that live in a specific directory. The scorecard is a Node parser that reads project data and calculates five metrics: discipline, rework ratio, time-to-ship, cost per shipped feature, and tokens-per-ship. The scorecard also includes a rule that if data isn't available, the metric is null, not estimated.
The author claims that projects using this protocol had significantly lower rework ratios and time-to-ship compared to those that didn't. The install script is straightforward and requires no dependencies beyond curl and git.
Written by urgent.news from Dev.to's reporting — not their text. Machine-written — may contain errors; check the original before relying on it.