Before you write a line of code, this runs multiple frontier models in parallel to sketch out types, signatures, and module boundaries with the callers written first. It grounds itself in your existing system using the 'how' skill, generates competing designs through 'arena', then synthesizes one approach and implements against it. The interesting bit is Phase E: if implementation keeps fighting the architecture, it throws everything out and redesigns from scratch rather than bolting on fixes. Default is fully autonomous, but you can add a checkpoint to review the sketch before it fills in the bodies. Best for non-trivial work where jumping straight to code would lock in the wrong shape.

skills.sh installs
5.6K
GitHub stars
15.9K
last updated
Aug 21, 2026

Install

$ npx -y skills add cursor/plugins --skill architect --agent claude-code

Installs into .claude/skills of the current project.

Files

Architect

Sketch types, function signatures, class shapes, and module boundaries with not implemented bodies and pseudocode. Synthesize across multiple model perspectives, then fill in code against the chosen sketch.

Start

Open a todolist with one entry per phase before starting.

  1. Ground
  2. Sketch
  3. Agree
  4. Implement
  5. Scrap

Phase A: Ground the problem

Build a real mental model of every system the new code touches. Run the how skill over the relevant subsystems. Naming a file isn't grounding. Produce the architectural explanation how prescribes. If the design redefines ownership or layering, also run the why skill on the existing shape so the rationale becomes a constraint, not a guess.

Skip Phase A only when the work is genuinely greenfield with no surrounding system to integrate.

Phase B: Sketch

Run the arena skill with the design-sketch task and the Phase A grounding artifacts. Pass references/runner-prompt.md as each runner's prompt.

Sponsors

Details

Repository
github.com/cursor/plugins
Category
Code Review & Quality
First seen
Jun 23, 2026

More from cursor/plugins

All 47
1softaworks avatarnaming-analyzersoftaworks/agent-toolkitSuggest better variable, function, and class names based on context and conventions.3.9Kskills.sh installs
2sickn33 avatarsoftware-architecturesickn33/antigravity-awesome-skillssoftware architecture3.3Kskills.sh installs
3lingzhi227 avatarself-reviewlingzhi227/agent-research-skillsAutomatically review an academic paper using the NeurIPS review form with three reviewer personas, ensemble scoring, and reflection refinement. Extracts text fr1.7Kskills.sh installs
4neolabhq avatarreview-local-changesneolabhq/context-engineering-kitReview your local uncommitted working-tree changes (git diff plus untracked files) and return actionable improvement suggestions. Use before committing, when no1.3Kskills.sh installs
5ruvnet avatargithub-code-reviewruvnet/rufloComprehensive GitHub code review with AI-powered swarm coordination1.3Kskills.sh installs
6ruvnet avataragent-code-review-swarmruvnet/rufloAgent skill for code-review-swarm - invoke with $agent-code-review-swarm1.1Kskills.sh installs