Cursor IDE Support

Everything Claude Code provides full Cursor IDE support with hooks, rules, agents, skills, commands, and MCP configs adapted for Cursor’s native format.

Overview

Cursor has more hook events than Claude Code (20 vs 8), providing richer automation capabilities. ECC leverages this with a DRY adapter pattern that reuses existing hook logic.

What’s Included

Component	Count	Details
Hook Events	15	sessionStart, beforeShellExecution, afterFileEdit, beforeMCPExecution, beforeSubmitPrompt, and 10 more
Hook Scripts	16	Thin Node.js scripts delegating to `scripts/hooks/` via shared adapter
Rules	29	9 common (alwaysApply) + 20 language-specific (TypeScript, Python, Go, Swift)
Agents	Shared	Via AGENTS.md at root (read by Cursor natively)
Skills	Shared + Bundled	Via AGENTS.md at root and `.cursor/skills/` for translated additions
Commands	Shared	`.cursor/commands/` if installed
MCP Config	Shared	`.cursor/mcp.json` if installed

Installation

Quick Install

# Install for your language(s)
./install.sh --target cursor typescript

# Multiple languages
./install.sh --target cursor python golang swift

The installer:

Copies rules to .cursor/rules/ with YAML frontmatter
Creates hook scripts in .cursor/hooks/
Sets up the DRY adapter pattern
Configures language-specific settings

Manual Installation

# Clone the repository
git clone https://github.com/affaan-m/everything-claude-code.git
cd everything-claude-code

# Copy rules (common + language-specific)
cp -r .cursor/rules/* ~/.cursor/rules/

# Copy hooks
cp -r .cursor/hooks/* ~/.cursor/hooks/

# AGENTS.md is auto-detected from repo root

Hook Architecture (DRY Adapter Pattern)

Cursor’s extensive hook system (20 events) is more powerful than Claude Code’s (8 events). The .cursor/hooks/adapter.js module transforms Cursor’s stdin JSON to Claude Code’s format, allowing existing scripts/hooks/*.js to be reused without duplication.

Cursor stdin JSON → adapter.js → transforms → scripts/hooks/*.js
                                               (shared with Claude Code)

Architecture Benefits

DRY Principle

Hook logic written once, works on both platforms

Zero Duplication

No need to maintain separate hook implementations

Cursor Advantages

More events = richer automation than Claude Code

Easy Maintenance

Bug fixes apply to both platforms automatically

Example Hook

const { transformCursorInput } = require('./adapter');
const shellCheck = require('../../scripts/hooks/shell-check');

const input = JSON.parse(require('fs').readFileSync(0, 'utf-8'));
const transformed = transformCursorInput('beforeShellExecution', input);
shellCheck(transformed);

// Shared hook logic used by both Cursor and Claude Code
module.exports = function(input) {
  const { command } = input;
  
  // Block dev servers outside tmux
  if (command.match(/npm run dev|yarn dev|pnpm dev/) && !process.env.TMUX) {
    console.error('[Hook] Dev servers must run in tmux');
    process.exit(2);  // Exit 2 = block execution
  }
  
  // Warn on git push
  if (command.startsWith('git push')) {
    console.warn('[Hook] Review changes before pushing');
  }
};

15 Hook Events

Cursor provides these hook events:

Session Lifecycle

sessionStart — Load context on session start
sessionEnd — Save state on session end
sessionIdle — Suggest compaction when idle

File Operations

beforeFileRead — Validate file access permissions
afterFileEdit — Auto-format, run type checks
beforeFileWrite — Block writes to protected files
beforeTabFileRead — Block Tab from reading .env, .key, .pem files

Shell Execution

beforeShellExecution — Block dev servers outside tmux, git push review
afterShellExecution — Log command results, detect failures

AI Interactions

beforeSubmitPrompt — Detect secrets (sk-, ghp_, AKIA) in prompts
afterPromptResponse — Extract learnings from responses

MCP Operations

beforeMCPExecution — Audit MCP tool calls
afterMCPExecution — Log MCP results

Other Events

beforeToolUse — Generic pre-tool validation
afterToolUse — Generic post-tool actions

Key Hooks in Action

# Blocks dev servers outside tmux (exit 2)
npm run dev  # ❌ Error: Dev servers must run in tmux

# Warns on git push
git push     # ⚠️  Warning: Review changes before pushing

# Auto-formats TypeScript files
# Runs TypeScript compiler check
# Warns if console.log detected

# Detects secrets in prompts
"Use API key sk-1234..."  # ❌ Error: Detected secret in prompt

# Blocks Tab from reading sensitive files (exit 2)
Tab tries to read .env   # ❌ Error: Cannot read .env file

Rules Format

Cursor rules use YAML frontmatter with description, globs, and alwaysApply:

---
description: "TypeScript coding style extending common rules"
globs: ["**/*.ts", "**/*.tsx", "**/*.js", "**/*.jsx"]
alwaysApply: false
---

# TypeScript Coding Standards

## Immutability

ALWAYS create new objects, NEVER mutate existing ones.

```typescript
// ❌ Bad - mutation
const user = { name: 'Alice' };
user.name = 'Bob';

// ✅ Good - immutable
const user = { name: 'Alice' };
const updated = { ...user, name: 'Bob' };

Type Safety

Use strict TypeScript settings. Avoid any.

### Common Rules (alwaysApply: true)

These apply to **all files** regardless of language:

- `coding-style.md` — Immutability, file organization
- `git-workflow.md` — Commit format, PR process
- `testing.md` — TDD, 80% coverage requirement
- `performance.md` — Model selection, context management
- `patterns.md` — Design patterns, skeleton projects
- `hooks.md` — Hook architecture, TodoWrite
- `agents.md` — When to delegate to subagents
- `security.md` — Mandatory security checks

### Language-Specific Rules (alwaysApply: false)

These activate based on file extension via `globs`:

<CodeGroup>
```yaml TypeScript
---
description: "TypeScript coding standards"
globs: ["**/*.ts", "**/*.tsx", "**/*.js", "**/*.jsx"]
alwaysApply: false
---

Python

---
description: "Python coding standards (PEP 8)"
globs: ["**/*.py"]
alwaysApply: false
---

---
description: "Go idioms and best practices"
globs: ["**/*.go"]
alwaysApply: false
---

Swift

---
description: "Swift coding standards"
globs: ["**/*.swift"]
alwaysApply: false
---

Configuration Example

.cursor/settings.json:

{
  "cursor.ai.model": "claude-sonnet-4.6",
  "cursor.ai.maxTokens": 10000,
  "cursor.ai.temperature": 0.7,
  "cursor.ai.enableHooks": true,
  "cursor.ai.rulesDirectory": ".cursor/rules"
}

Feature Comparison

Feature	Claude Code	Cursor IDE
Agents	13 agents	Shared (AGENTS.md)
Commands	33 commands	Shared
Skills	50+ skills	Shared + Bundled
Hook Events	8 types	15 types ✨
Hook Scripts	9 scripts	16 scripts
Rules	29 rules	29 rules (YAML)
Custom Tools	Via hooks	Via hooks
MCP Servers	14 servers	Shared
Secret Detection	Hook-based	beforeSubmitPrompt hook ✨
Auto-Format	PostToolUse hook	afterFileEdit hook ✨

Cursor Advantages: More hook events, richer automation, native IDE integration

Usage Examples

Starting a Feature

# Cursor detects AGENTS.md automatically
# Just start prompting:

"Plan authentication with OAuth"
# → planner agent creates blueprint

"Implement with TDD"
# → tdd-guide enforces write-tests-first

"Review my changes"
# → code-reviewer checks quality

Auto-Formatting on Save

Cursor’s afterFileEdit hook automatically:

Runs Prettier on TypeScript/JavaScript files
Runs TypeScript compiler check
Warns if console.log detected
Updates documentation if needed

No manual action required!

Secret Detection

# Try to submit a prompt with a secret:
"Deploy with API key sk-1234567890"

# beforeSubmitPrompt hook blocks it:
# ❌ Error: Detected secret pattern in prompt (sk-)

Best Practices

Use AGENTS.md

Keep project-specific context in AGENTS.md at repo root

Leverage Hooks

Cursor’s 15 hook events enable rich automation

Language Rules

Install only rules for languages you use

Test Hooks

Verify hooks work by triggering events manually

Troubleshooting

Hooks not firing

Check .cursor/hooks/ exists and scripts are executable
Verify cursor.ai.enableHooks is true in settings
Check hook script exit codes (0 = allow, 2 = block)
View Cursor logs for hook execution errors

Rules not applying

Verify rules are in .cursor/rules/
Check YAML frontmatter is valid
Ensure globs match your file extensions
Common rules need alwaysApply: true

AGENTS.md not detected

Place AGENTS.md at repository root
Restart Cursor
Open a file in the project
Check Cursor AI panel for context loading

Next Steps

Hook Examples

See all 16 hook scripts in detail

Rules Guide

Learn about the 29 rules and how to customize

Agents

Explore shared agent system

Skills

Browse 50+ skills available in Cursor

Get Started

Core Concepts

Platform Support

Guides

Overview

What’s Included

Installation

Quick Install

Manual Installation

Hook Architecture (DRY Adapter Pattern)

Architecture Benefits

DRY Principle

Zero Duplication

Cursor Advantages

Easy Maintenance

Example Hook

15 Hook Events

Key Hooks in Action

Rules Format

Type Safety

Configuration Example

Feature Comparison

Usage Examples

Starting a Feature

Auto-Formatting on Save

Secret Detection

Best Practices

Use AGENTS.md

Leverage Hooks

Language Rules

Test Hooks

Troubleshooting

Next Steps

Hook Examples

Rules Guide

Agents

Skills

​Overview

​What’s Included

​Installation

​Quick Install

​Manual Installation

​Hook Architecture (DRY Adapter Pattern)

​Architecture Benefits

DRY Principle

Zero Duplication

Cursor Advantages

Easy Maintenance

​Example Hook

​15 Hook Events

​Key Hooks in Action

​Rules Format

​Type Safety

​Configuration Example

​Feature Comparison

​Usage Examples

​Starting a Feature

​Auto-Formatting on Save

​Secret Detection

​Best Practices

Use AGENTS.md

Leverage Hooks

Language Rules

Test Hooks

​Troubleshooting

​Next Steps

Hook Examples

Rules Guide

Agents

Skills

Overview

What’s Included

Installation

Quick Install

Manual Installation

Hook Architecture (DRY Adapter Pattern)

Architecture Benefits

Example Hook

15 Hook Events

Key Hooks in Action

Rules Format

Type Safety

Configuration Example

Feature Comparison

Usage Examples

Starting a Feature

Auto-Formatting on Save

Secret Detection

Best Practices

Troubleshooting

Next Steps