Claude to cursor

Skill scoobydont-666/shared-claude-skills/skills/claude-to-cursor

10 production-tested Claude Code skills — model routing, security hardening, code quality, tax advisory, cost optimization. Install: clone to ~/.claude/skills/

Install

npx -y skills add scoobydont-666/shared-claude-skills --skill claude-to-cursor

Assembled from the repository path, not quoted from the project. Check it against their README if it does not work.

One thing to look at

1 stars1 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.

What its author says it does

Copied from the file, not written here

Convert Claude Code skills to Cursor-compatible rules (.mdc/.md files) for sharing or use in Cursor workspaces. Does NOT modify local Claude Code setup — outputs go to a separate repo. Trigger on: "convert to cursor", "cursor rules", "port to cursor", ".mdc", "cursor format", or any request to export skills for Cursor users.

SKILL.md

3.3 KB, as published. Nobody here has run it

Claude to Cursor Converter

Export Claude Code skills as Cursor-compatible rule files for sharing. All output goes to a dedicated repo — never modifies local Claude Code configuration.

Cursor Rule Format (.mdc)

Cursor uses .mdc files (Markdown with frontmatter) in .cursor/rules/:

---
description: One-line description (used for rule matching)
globs:
  - "**/*.py"          # optional: file patterns this rule applies to
alwaysApply: false      # true = always loaded, false = matched by description
---

# Rule Title

Rule content here. Markdown format.

Key differences from Claude SKILL.md:

Extension: .mdc (not .md — Cursor's convention for rules)
No name: field — filename is the identifier
globs: for file-pattern matching (Claude uses description triggers)
alwaysApply: replaces Claude's always-loaded behavior
No references/ subdirectory convention — inline or split into companion rules

Conversion Process

1. Classify the skill

Type	Converts?	Notes
Persona (infosec-architect, cpa-tax-specialist)	Yes	Direct conversion, strong match
Workflow (tdd, write-a-prd, prd-to-plan)	Yes	Direct conversion
Code quality (code-consistency, code-quality)	Yes	Use globs for language-specific rules
Tool-specific (token-miser, budi-analytics)	No	Claude Code-only features
Infrastructure (fleet-manager, openclaw-ops)	Partial	Remove Claude-specific tool refs

2. Transform content

Strip YAML frontmatter, rewrite as Cursor frontmatter (description, globs, alwaysApply)
Remove Claude-specific tool references (Agent tool, Bash tool, Read/Write/Edit, MCP)
Inline small reference files (<200 lines) directly
Large references → separate .mdc companion files
Replace "Claude Code" with generic "AI assistant" where contextual
Add attribution:

3. Output structure

claude-to-cursor/
├── rules/                    # Converted .mdc files
│   ├── tdd.mdc
│   ├── tdd-tests-ref.mdc    # Large reference companion
│   ├── code-quality.mdc
│   └── ...
├── agents/                   # Converted agent files
│   └── ...
└── scripts/
    └── convert.sh            # Batch conversion script

4. What to skip

token-miser (Claude Code subagent routing — no Cursor equivalent)
budi-analytics (Claude Code hook analytics — Cursor-specific)
skill-updater (meta-skill for Claude Code skill management)
inbound-sync (claude.ai ↔ Claude Code sync workflow)
openclaw-ops (primary-host-specific deployment)
fleet-manager (SSH/Ansible fleet ops — not IDE-relevant)
ansible-hardening (infrastructure, not IDE-relevant)

Rules

NEVER modify local Claude Code skills or configuration
All output goes to the claude-to-cursor repo only
Keep attribution on all converted files
Test in Cursor before marking as ready
.mdc extension for rules, .md for agents