A curated collection of AI coding rules, guidelines, agent definitions, and skills for use with Claude and Cursor.
This repository organizes reusable rule files, agent definitions, and skills by category, helping AI assistants generate consistent, high-quality code and perform specialized workflows across different projects.
.
├── Claude/ # Claude-specific resources
│ ├── agents/ # Sub-agent definitions (.md with YAML frontmatter)
│ ├── rules/ # Coding standards and guidelines
│ │ ├── common/ # Language-agnostic principles (always install)
│ │ ├── typescript/
│ │ ├── python/
│ │ ├── golang/
│ │ ├── web/
│ │ ├── swift/
│ │ ├── php/
│ │ └── ... # Other language-specific rule sets
│ └── skills/ # Agent skills (SKILL.md format)
│ ├── create-pr/
│ ├── pr-review/
│ └── ... # 180+ skills across domains
└── Cursor/ # Cursor .mdc rules by technology stack
├── backend/
│ ├── db/
│ │ └── postgresql.mdc
│ └── python/
│ ├── boto3.mdc
│ ├── django.mdc
│ ├── django-rest-framework.mdc
│ ├── fastapi.mdc
│ ├── flask.mdc
│ ├── flask-restful.mdc
│ ├── pydantic.mdc
│ └── sqlalchemy.mdc
├── code-as-infra/
│ ├── docker.mdc
│ └── terraform.mdc
└── frontend/
├── next-js.mdc
├── react.mdc
└── three-js.mdc
Agent definitions live in Claude/agents/. Each file is a Markdown document with YAML frontmatter specifying the agent's name, description, available tools, and preferred model.
Copy the agent files you need into ~/.claude/agents/ (global) or .claude/agents/ (project-level).
Example agents:
| Agent | Description |
|---|---|
architect |
Software architecture specialist for system design and technical decisions |
code-reviewer |
Expert code review for quality, security, and maintainability |
security-reviewer |
Security-focused review for vulnerabilities and compliance |
python-reviewer |
Python-specific code quality and idiom checks |
tdd-guide |
Test-driven development guidance and workflow |
performance-optimizer |
Performance analysis and optimization recommendations |
47 agents are available. Browse
Claude/agents/for the full list.
Rules live in Claude/rules/ and are organized into a common layer plus language-specific directories. See Claude/rules/README.md for full installation instructions and details on rule priority.
Manual install:
# Common rules are required for all projects
cp -r Claude/rules/common ~/.claude/rules/common
# Add language-specific rules as needed
cp -r Claude/rules/python ~/.claude/rules/python
cp -r Claude/rules/typescript ~/.claude/rules/typescript
cp -r Claude/rules/golang ~/.claude/rules/golangAvailable rule sets: common, typescript, python, golang, web, swift, php, cpp, csharp, dart, java, kotlin, perl, rust
Skills live in Claude/skills/. Each skill is a directory containing a SKILL.md file that teaches the agent a specialized workflow.
Copy the skill directories you need into ~/.claude/skills/ (global) or .claude/skills/ (project-level).
Selected skills:
| Skill | Description |
|---|---|
| create-pr | Create a GitHub PR from the current branch using the gh CLI |
| pr-review | Review GitHub PRs and post inline code review comments |
| django-patterns | Django best practices and patterns |
| python-testing | Python testing workflows and patterns |
| security-review | Security review checklist and workflow |
| tdd-workflow | Test-driven development workflow |
| git-workflow | Git branching, committing, and collaboration workflow |
| api-design | REST and API design guidelines |
183 skills are available across domains including backend, frontend, DevOps, AI/ML, and more. Browse
Claude/skills/for the full list.
Copy the relevant .mdc rule files into your project's .cursor/rules/ directory. Refer to the included Guidelines for Applying Cursor Rules.pdf for best practices on structuring and applying rules effectively.
Available rule files:
| Category | Rules |
|---|---|
| Backend / Python | django, django-rest-framework, fastapi, flask, flask-restful, boto3, pydantic, sqlalchemy |
| Backend / Database | postgresql |
| Code as Infrastructure | docker, terraform |
| Frontend | next-js, react, three-js |
See CONTRIBUTING.md for full guidelines. Pull requests are welcome!