Jup Architect
A real, well-grounded architecture-review rubric for the jup CLI's own codebase — but it only means anything if you're literally working inside the jup repo.
Test report
- Verdict
- Works with setup
- Score
- Tested
- Jul 16, 2026
- Environment
- Claude Code 2.x (agent harness)
- Upstream re-checked
- Aug 10, 2026 · 562191c
Used as intended, inside the real jup repo, its Data Persistence focus area guided a real find: get_skills_lock() in src/jup/config.py silently swallows any parse exception and returns an empty SkillsLock(), which skills_lock_session() then persists back to disk on exit — a traceable path to silently wiping a user's real lockfile on any transient corruption. But the skill's own Deliverables section points at docs/superpowers/specs/ and docs/superpowers/plans/, directories that don't exist and aren't part of jup's actual documented workflow (AGENTS.md specifies backlog/roadmap.md instead) — stale copy-paste from a template, and outside the jup repo itself this skill is functionally inert.
Scored on four weighted criteria — install, triggering, output vs. baseline, docs. How scoring works
- Installs cleanly 5/5
- Triggers reliably 5/5
- Output vs. baseline 8/10
- Docs & honesty 3/5
What Jup Architect does
An architecture-review checklist scoped entirely to the `jup` skill-sync CLI's own codebase (src/jup/commands, config.py, models.py, the ~/.jup lockfile) — covering modularity, idiomatic Typer usage, Pydantic type safety, error handling, and testability, with a process flow and deliverable targets. Triggers on requests to audit jup's technical debt or propose architectural changes for jup specifically; not a general-purpose codebase-review skill.
How to install Jup Architect
git clone https://github.com/andrader/jup
cd jup
mkdir -p ~/.claude/skills
cp -r skills/jup-architect ~/.claude/skills/jup-architect
Skills live in ~/.claude/skills/ (global) or .claude/skills/
(per-project). Restart Claude Code after installing.
Commands — how to trigger Jup Architect
-
/jup-architectA real, well-grounded architecture-review rubric for the jup CLI's own codebase — but it only means anything if you're literally working inside the jup repo.
It also activates on plain-language prompts like these:
-
Review src/jup/config.py and models.py for architecture issues -
Audit jup's lockfile persistence logic for silent data loss bugs -
Propose architectural changes for jup's command modularity
Frequently asked questions
- Is the Jup Architect skill free?
- Yes. The skill itself is free from andrader/jup. SkillProof publishes the install command and an independent test verdict at no cost.
- Does Jup Architect work with Claude Code?
- We tested it with Claude Code 2.x (agent harness) on Jul 16, 2026. Verdict: Works with setup. Used as intended, inside the real jup repo, its Data Persistence focus area guided a real find: get_skills_lock() in src/jup/config.py silently swallows any parse exception and returns an empty SkillsLock(), which skills_lock_session() then persists back to disk on exit — a traceable path to silently wiping a user's real lockfile on any transient corruption. But the skill's own Deliverables section points at docs/superpowers/specs/ and docs/superpowers/plans/, directories that don't exist and aren't part of jup's actual documented workflow (AGENTS.md specifies backlog/roadmap.md instead) — stale copy-paste from a template, and outside the jup repo itself this skill is functionally inert.
- What is the Jup Architect SkillProof Score?
- 8.4/10 — installs cleanly 5/5, triggers reliably 5/5, output vs. baseline 8/10, docs & honesty 3/5.
- How do I install Jup Architect?
- Copy the install command from this page, run it in your terminal, and restart Claude Code. Skills live in ~/.claude/skills/ (global) or .claude/skills/ inside a project.
- Can I use Jup Architect with Cursor, Copilot, Gemini CLI, Codex or other AI tools?
- The SKILL.md format is native to Claude (Claude Code, Desktop, claude.ai). The instructions inside adapt to other assistants: Cursor rules, GitHub Copilot instructions, Windsurf rules, Custom GPTs, AGENTS.md for OpenAI Codex, and GEMINI.md for Google Gemini CLI — our conversion guides cover each, and the free converter on the tools page does the wrapping for you.