Agnostique/uprompt
Turn a plain-language note into a disciplined Claude Code task — read, fix, verify with tests not screenshots, commit. Installable plugin, domain-agnostic.
SUMMARY AI summary by gpt-5-mini
uprompt provides a disciplined "/up" command for Claude Code and other Claude-based coding agents that turns casual, plain-language requests into concrete engineering tasks. You invoke /up (e.g. "/up fix checkout button overlap") and the agent: reads the repo, diagnoses root cause, applies minimal code changes, runs automated verification (tests/logs/build — explicitly no screenshots), creates one conventional commit (never pushes), and reports what to eyeball. Key features: protected-zone declarations to forbid edits to sensitive/generated files, a "solved — don't redo" section to avoid regressions, configurable verify commands, and safe-by-default behavior (no deploys or force-pushes). Installable by copying commands/up.md into .claude or via a plugin; must fill project-specific placeholders. Target audience: solo devs and small teams using Claude Code who want repeatable, reviewable fixes. Licensed PolyForm Noncommercial 1.0.0.
DETECTED Detected AI stack
AI-related keywords found in this repo's description, topics, or README summary — grouped by category. Each badge links to the corresponding ranking detail page.
GitHub Topics
Owner
Dates
| Created on GitHub | 2026-05-31 |
| Last push | 2026-05-31 |
| First seen here | 2026-05-31 |
| Last fetched | 2026-05-31 16:10 |