alirezarezvani/claude-code-skill-factory / generated-skills/hook-factory
hook-factory
FlaggedGenerate production-ready Claude Code hooks with interactive Q&A, automated installation, and enhanced validation. Supports 10 templates across 7 event types for comprehensive workflow automation.
templates.json
View on GitHub{
"post_tool_use_format": {
"metadata": {
"name": "Auto-Format Code After Editing",
"description": "Automatically format code files after editing with language-specific formatters",
"use_cases": [
"Auto-format Python files with black",
"Auto-format JavaScript/TypeScript with prettier",
"Auto-format Rust with rustfmt"
],
"complexity": "beginner",
"safety_notes": "Must complete in <5 seconds. Uses silent failure mode.",
"event_type": "PostToolUse",
"timing": "Triggers immediately after Write/Edit tools complete"
},
"template": {
"matcher": {
"tool_names": ["Write", "Edit"]
},
"hooks": [
{
"type": "command",
"command": "# Check if {{TOOL}} is installed\nif ! command -v {{TOOL}} &> /dev/null; then\n exit 0\nfi\n\n# Format the file based on type\n{{FORMAT_LOGIC}}",
"timeout": 60
}
]
},
"variables": {
"TOOL": {
"description": "Formatting tool to use",
"options": {
"python": "black",
"javascript": "prettier",
"typescript": "prettier",
"rust": "rustfmt",
"go": "gofmt"
}
},
"FORMAT_LOGIC": {
"description": "Language-specific formatting logic",
"patterns": {
"python": "if [[ \"$CLAUDE_TOOL_FILE_PATH\" == *.py ]]; then\n black \"$CLAUDE_TOOL_FILE_PATH\" || exit 0\nfi",
"javascript": "if [[ \"$CLAUDE_TOOL_FILE_PATH\" == *.js ]] || [[ \"$CLAUDE_TOOL_FILE_PATH\" == *.jsx ]]; then\n prettier --write \"$CLAUDE_TOOL_FILE_PATH\" || exit 0\nfi",
"typescript": "if [[ \"$CLAUDE_TOOL_FILE_PATH\" == *.ts ]] || [[ \"$CLAUDE_TOOL_FILE_PATH\" == *.tsx ]]; then\n prettier --write \"$CLAUDE_TOOL_FILE_PATH\" || exit 0\nfi",
"rust": "if [[ \"$CLAUDE_TOOL_FILE_PATH\" == *.rs ]]; then\n rustfmt \"$CLAUDE_TOOL_FILE_PATH\" || exit 0\nfi",
"go": "if [[ \"$CLAUDE_TOOL_FILE_PATH\" == *.go ]]; then\n gofmt -w \"$CLAUDE_TOOL_FILE_PATH\" || exit 0\nfi"
}
},
"FILE_PATTERNS": {
"description": "Glob patterns for target files",
"options": {
"python": ["**/*.py"],
"javascript": ["**/*.js", "**/*.jsx"],
"typescript": ["**/*.ts", "**/*.tsx"],
"rust": ["**/*.rs"],
"go": ["**/*.go"]
}
}
}
},
"post_tool_use_git_add": {
"metadata": {
"name": "Auto-Add Files to Git After Editing",
"description": "Automatically stage modified files with git after editing",
"use_cases": [
"Auto-stage all edited files",
"Auto-stage specific file types",
"Simplify git workflow"
],
"complexity": "beginner",
"safety_notes": "Read git status before staging. Uses silent failure mode.",
"event_type": "PostToolUse",
"timing": "Triggers immediately after Write/Edit tools complete"
},
"template": {
"matcher": {
"tool_names": ["Write", "Edit"]
},
"hooks": [
{
"type": "command",
"command": "# Check if this is a git repository\nif ! git rev-parse --git-dir &> /dev/null; then\n exit 0\nfi\n\n# Check if file path is available\nif [ -z \"$CLAUDE_TOOL_FILE_PATH\" ]; then\n exit 0\nfi\n\n# Stage the file\ngit add \"$CLAUDE_TOOL_FILE_PATH\" 2>/dev/null || exit 0",
"timeout": 60
}
]
},
"variables": {
"FILE_FILTER": {
"description": "Optional filter for which files to auto-add",
"options": {
"all": "Stage all edited files",
"code_only": "Stage only code files (*.py, *.js, *.ts, etc.)",
"exclude_tests": "Stage all except test files"
},
"default": "all"
}
}
},
"subagent_stop_test_runner": {
"metadata": {
"name": "Run Tests When Agent Completes",
"description": "Automatically run tests after agent completes work",
"use_cases": [
"Run unit tests after code changes",
"Run integration tests",
"Quality gate before continuing"
],
"complexity": "intermediate",
"safety_notes": "Can be slower (tests take time). Runs once per agent completion.",
"event_type": "SubagentStop",
"timing": "Triggers once when agent finishes its task"
},
"template": {
"matcher": {},
"hooks": [
{
"type": "command",
"command": "# Detect test framework and run tests\n{{TEST_DETECTION_LOGIC}}\n\n# Run the appropriate test command\n{{TEST_COMMAND}} || exit 0",
"timeout": 120
}
]
},
"variables": {
"TEST_FRAMEWORK": {
"description": "Testing framework to use",
"options": {
"python": "pytest",
"javascript": "jest",
"typescript": "jest",
"rust": "cargo",
"go": "go"
}
},
"TEST_DETECTION_LOGIC": {
"description": "Logic to detect which test framework to use",
"patterns": {
"python": "if ! command -v pytest &> /dev/null; then\n exit 0\nfi",
"javascript": "if ! command -v npm &> /dev/null; then\n exit 0\nfi\nif [ ! -f package.json ]; then\n exit 0\nfi",
"rust": "if ! command -v cargo &> /dev/null; then\n exit 0\nfi",
"go": "if ! command -v go &> /dev/null; then\n exit 0\nfi"
}
},
"TEST_COMMAND": {
"description": "Command to run tests",
"patterns": {
"python": "pytest -v",
"javascript": "npm test",
"rust": "cargo test",
"go": "go test ./..."
}
}
}
},
"session_start_context_loader": {
"metadata": {
"name": "Load Project Context at Session Start",
"description": "Load project-specific context when Claude Code session starts",
"use_cases": [
"Load TODO list",
"Display project status",
"Show recent git changes",
"Load custom project instructions"
],
"complexity": "beginner",
"safety_notes": "Read-only operations recommended. Fast execution required.",
"event_type": "SessionStart",
"timing": "Triggers once when session begins or resumes"
},
"template": {
"matcher": {
"source": ["startup", "resume"]
},
"hooks": [
{
"type": "command",
"command": "# Load project context\n{{CONTEXT_LOAD_LOGIC}}",
"timeout": 60
}
]
},
"variables": {
"CONTEXT_FILE": {
"description": "File to load for context",
"options": {
"todo": "TODO.md",
"agents": "AGENTS.md",
"project_status": "PROJECT_STATUS.md",
"custom": "CONTEXT.md"
}
},
"CONTEXT_LOAD_LOGIC": {
"description": "Logic to load and display context",
"patterns": {
"todo": "if [ -f TODO.md ]; then\n echo \"# Project TODO List\"\n cat TODO.md\nelse\n echo \"No TODO.md found\"\nfi",
"agents": "if [ -f AGENTS.md ]; then\n echo \"# Agent Configuration\"\n cat AGENTS.md\nelse\n echo \"No AGENTS.md found\"\nfi",
"project_status": "if [ -f PROJECT_STATUS.md ]; then\n echo \"# Project Status\"\n cat PROJECT_STATUS.md\nelse\n echo \"No PROJECT_STATUS.md found\"\nfi",
"git_status": "if git rev-parse --git-dir &> /dev/null; then\n echo \"# Git Status\"\n git status --short\nelse\n echo \"Not a git repository\"\nfi"
}
}
}
},
"pre_tool_use_validation": {
"metadata": {
"name": "Validate Before Tool Execution",
"description": "Validate inputs and permissions before tool execution to prevent errors",
"use_cases": [
"Check file permissions before Write/Edit",
"Validate paths exist before operations",
"Prevent destructive operations on protected files",
"Verify disk space before large writes"
],
"complexity": "intermediate",
"safety_notes": "BLOCKING event - must complete in <10 seconds. Can prevent tool from executing.",
"event_type": "PreToolUse",
"timing": "Triggers BEFORE tool executes - can block operation"
},
"template": {
"matcher": {
"tool_names": ["Write", "Edit", "Bash"]
},
"hooks": [
{
"type": "command",
"command": "# Validate before tool execution\n{{VALIDATION_LOGIC}}",
"timeout": 10
}
]
},
"variables": {
"VALIDATION_TYPE": {
"description": "Type of validation to perform",
"options": {
"permissions": "Check file write permissions",
"path_exists": "Validate path exists for reads",
"disk_space": "Check available disk space",
"protected_files": "Prevent operations on protected files"
}
},
"VALIDATION_LOGIC": {
"description": "Validation logic to execute",
"patterns": {
"permissions": "if [ -n \"$CLAUDE_TOOL_FILE_PATH\" ] && [ -f \"$CLAUDE_TOOL_FILE_PATH\" ]; then\n if [ ! -w \"$CLAUDE_TOOL_FILE_PATH\" ]; then\n echo \"Error: No write permission for $CLAUDE_TOOL_FILE_PATH\"\n exit 1\n fi\nfi",
"path_exists": "if [ -n \"$CLAUDE_TOOL_FILE_PATH\" ] && [ ! -e \"$CLAUDE_TOOL_FILE_PATH\" ]; then\n PARENT_DIR=$(dirname \"$CLAUDE_TOOL_FILE_PATH\")\n if [ ! -d \"$PARENT_DIR\" ]; then\n echo \"Error: Parent directory does not exist: $PARENT_DIR\"\n exit 1\n fi\nfi",
"disk_space": "REQUIRED_SPACE=100000 # 100MB in KB\nAVAILABLE=$(df . | tail -1 | awk '{print $4}')\nif [ \"$AVAILABLE\" -lt \"$REQUIRED_SPACE\" ]; then\n echo \"Error: Insufficient disk space (need 100MB)\"\n exit 1\nfi",
"protected_files": "if [ -n \"$CLAUDE_TOOL_FILE_PATH\" ]; then\n PROTECTED_PATTERNS=(\"/.env$\" \"/secrets\" \"/credentials\" \"/.aws/\" \"/.ssh/\")\n for pattern in \"${PROTECTED_PATTERNS[@]}\"; do\n if [[ \"$CLAUDE_TOOL_FILE_PATH\" =~ $pattern ]]; then\n echo \"Error: Cannot modify protected file: $CLAUDE_TOOL_FILE_PATH\"\n exit 1\n fi\n done\nfi"
}
}
}
},
"user_prompt_submit_preprocessor": {
"metadata": {
"name": "Pre-process User Prompts",
"description": "Pre-process and validate user prompts before Claude processes them",
"use_cases": [
"Add project context to prompts automatically",
"Validate prompt safety before processing",
"Inject custom instructions",
"Log user prompts for analysis"
],
"complexity": "advanced",
"safety_notes": "BLOCKING event - must complete in <5 seconds. Can modify or reject prompts.",
"event_type": "UserPromptSubmit",
"timing": "Triggers BEFORE Claude processes user input - can block processing"
},
"template": {
"matcher": {},
"hooks": [
{
"type": "command",
"command": "# Pre-process user prompt\n{{PREPROCESS_LOGIC}}",
"timeout": 5
}
]
},
"variables": {
"PREPROCESS_TYPE": {
"description": "Type of pre-processing to perform",
"options": {
"add_context": "Inject project context into prompts",
"validate_safety": "Check for potentially unsafe requests",
"log_prompts": "Log prompts to file for analysis",
"inject_instructions": "Add custom instructions to prompts"
}
},
"PREPROCESS_LOGIC": {
"description": "Pre-processing logic to execute",
"patterns": {
"add_context": "# Add project context to prompt\nif [ -f .claude/context.txt ]; then\n echo \"[Context: $(cat .claude/context.txt)]\"\nfi",
"validate_safety": "# Check for unsafe patterns\nif [ -n \"$CLAUDE_USER_PROMPT\" ]; then\n UNSAFE_PATTERNS=(\"rm -rf /\" \"sudo rm\" \"format disk\")\n for pattern in \"${UNSAFE_PATTERNS[@]}\"; do\n if [[ \"$CLAUDE_USER_PROMPT\" =~ $pattern ]]; then\n echo \"Warning: Potentially unsafe request detected\"\n fi\n done\nfi",
"log_prompts": "# Log prompt to file\nif [ -n \"$CLAUDE_USER_PROMPT\" ]; then\n echo \"[$(date '+%Y-%m-%d %H:%M:%S')] $CLAUDE_USER_PROMPT\" >> .claude/prompt_log.txt\nfi",
"inject_instructions": "# Inject custom instructions\nif [ -f .claude/custom_instructions.txt ]; then\n echo \"[Instructions: $(cat .claude/custom_instructions.txt)]\"\nfi"
}
}
}
},
"stop_session_cleanup": {
"metadata": {
"name": "Session Cleanup on Exit",
"description": "Clean up resources and save state when Claude Code session ends",
"use_cases": [
"Clean up temporary files",
"Save session state",
"Generate session summary report",
"Backup important files"
],
"complexity": "beginner",
"safety_notes": "Non-blocking. Should complete in <30 seconds. Safe cleanup operations only.",
"event_type": "Stop",
"timing": "Triggers once when session ends normally"
},
"template": {
"matcher": {},
"hooks": [
{
"type": "command",
"command": "# Session cleanup\n{{CLEANUP_LOGIC}}",
"timeout": 30
}
]
},
"variables": {
"CLEANUP_TYPE": {
"description": "Type of cleanup to perform",
"options": {
"temp_files": "Remove temporary files",
"save_state": "Save session state to file",
"generate_summary": "Generate session summary",
"backup_files": "Backup modified files"
}
},
"CLEANUP_LOGIC": {
"description": "Cleanup logic to execute",
"patterns": {
"temp_files": "# Clean up temporary files\nfind /tmp -name 'claude-*' -mtime +1 -delete 2>/dev/null || exit 0\nfind . -name '*.tmp' -delete 2>/dev/null || exit 0",
"save_state": "# Save session state\nif [ -d .claude ]; then\n echo \"Session ended: $(date)\" >> .claude/session_history.txt\n git status --short >> .claude/session_history.txt 2>/dev/null || exit 0\nfi",
"generate_summary": "# Generate session summary\nif git rev-parse --git-dir &> /dev/null; then\n echo \"# Session Summary - $(date)\" > .claude/session_summary.md\n echo \"## Files Modified\" >> .claude/session_summary.md\n git diff --name-only >> .claude/session_summary.md 2>/dev/null || exit 0\nfi",
"backup_files": "# Backup modified files\nif git rev-parse --git-dir &> /dev/null; then\n BACKUP_DIR=\".claude/backups/$(date +%Y%m%d_%H%M%S)\"\n mkdir -p \"$BACKUP_DIR\"\n git diff --name-only | while read file; do\n if [ -f \"$file\" ]; then\n cp \"$file\" \"$BACKUP_DIR/\" 2>/dev/null || exit 0\n fi\n done\nfi"
}
}
}
},
"pre_push_validation": {
"metadata": {
"name": "Validate Before Git Push",
"description": "Run validation checks before git push to ensure code quality",
"use_cases": [
"Run tests before push",
"Check commit messages",
"Validate branch protection rules",
"Run linters and formatters"
],
"complexity": "intermediate",
"safety_notes": "BLOCKING event - can take 60-120 seconds. Can prevent push if validation fails.",
"event_type": "PrePush",
"timing": "Triggers BEFORE git push executes - can block push"
},
"template": {
"matcher": {},
"hooks": [
{
"type": "command",
"command": "# Pre-push validation\n{{VALIDATION_LOGIC}}",
"timeout": 120
}
]
},
"variables": {
"VALIDATION_TYPE": {
"description": "Type of validation to perform",
"options": {
"run_tests": "Run test suite before push",
"check_commits": "Validate commit messages",
"branch_protection": "Check branch protection rules",
"lint_code": "Run linters and formatters"
}
},
"VALIDATION_LOGIC": {
"description": "Validation logic to execute",
"patterns": {
"run_tests": "# Run tests before push\nif [ -f package.json ]; then\n echo \"Running tests...\"\n npm test || { echo \"Tests failed - push blocked\"; exit 1; }\nelif [ -f pytest.ini ] || [ -f setup.py ]; then\n echo \"Running tests...\"\n pytest || { echo \"Tests failed - push blocked\"; exit 1; }\nelse\n echo \"No test framework detected - skipping\"\nfi",
"check_commits": "# Validate commit messages\nCURRENT_BRANCH=$(git branch --show-current)\nREMOTE_BRANCH=\"origin/$CURRENT_BRANCH\"\nif git rev-parse \"$REMOTE_BRANCH\" &> /dev/null; then\n COMMITS=$(git log \"$REMOTE_BRANCH\"..HEAD --oneline)\n if echo \"$COMMITS\" | grep -qE '^[0-9a-f]+ (WIP|TODO|FIXME)'; then\n echo \"Error: Commit messages contain WIP/TODO/FIXME\"\n exit 1\n fi\nfi",
"branch_protection": "# Check branch protection\nCURRENT_BRANCH=$(git branch --show-current)\nPROTECTED_BRANCHES=(\"main\" \"master\" \"production\")\nfor branch in \"${PROTECTED_BRANCHES[@]}\"; do\n if [ \"$CURRENT_BRANCH\" = \"$branch\" ]; then\n echo \"Error: Direct push to protected branch '$branch' is not allowed\"\n exit 1\n fi\ndone",
"lint_code": "# Run linters\nif [ -f .eslintrc.js ] || [ -f .eslintrc.json ]; then\n echo \"Running ESLint...\"\n npx eslint . || { echo \"Linting failed - push blocked\"; exit 1; }\nelif command -v pylint &> /dev/null; then\n echo \"Running pylint...\"\n pylint **/*.py || { echo \"Linting failed - push blocked\"; exit 1; }\nelse\n echo \"No linter detected - skipping\"\nfi"
}
}
}
},
"notify_user_desktop": {
"metadata": {
"name": "Desktop Notifications",
"description": "Send desktop notifications for important events (macOS/Linux)",
"use_cases": [
"Notify when agent completes",
"Alert on test failures",
"Notify on successful operations",
"Sound alerts for long-running tasks"
],
"complexity": "beginner",
"safety_notes": "Non-blocking. Fast execution (<5 seconds). macOS uses osascript, Linux uses notify-send.",
"event_type": "PostToolUse",
"timing": "Triggers after tool completes or agent finishes"
},
"template": {
"matcher": {
"tool_names": ["Bash", "Task"]
},
"hooks": [
{
"type": "command",
"command": "# Send desktop notification\n{{NOTIFICATION_LOGIC}}",
"timeout": 5
}
]
},
"variables": {
"NOTIFICATION_TYPE": {
"description": "Type of notification to send",
"options": {
"agent_complete": "Notify when agent completes",
"test_result": "Notify test pass/fail",
"operation_complete": "Notify on operation completion",
"sound_alert": "Play sound alert"
}
},
"NOTIFICATION_LOGIC": {
"description": "Notification logic for macOS/Linux",
"patterns": {
"agent_complete": "# Notify agent completion\nif [[ \"$OSTYPE\" == \"darwin\"* ]]; then\n osascript -e 'display notification \"Agent task completed\" with title \"Claude Code\" sound name \"Glass\"'\nelif command -v notify-send &> /dev/null; then\n notify-send \"Claude Code\" \"Agent task completed\"\nfi",
"test_result": "# Notify test results\nif [[ \"$OSTYPE\" == \"darwin\"* ]]; then\n if [ $? -eq 0 ]; then\n osascript -e 'display notification \"All tests passed\" with title \"Test Results\" sound name \"Glass\"'\n else\n osascript -e 'display notification \"Tests failed\" with title \"Test Results\" sound name \"Basso\"'\n fi\nelif command -v notify-send &> /dev/null; then\n if [ $? -eq 0 ]; then\n notify-send \"Test Results\" \"All tests passed\"\n else\n notify-send -u critical \"Test Results\" \"Tests failed\"\n fi\nfi",
"operation_complete": "# Notify operation complete\nif [[ \"$OSTYPE\" == \"darwin\"* ]]; then\n osascript -e 'display notification \"Operation completed successfully\" with title \"Claude Code\"'\nelif command -v notify-send &> /dev/null; then\n notify-send \"Claude Code\" \"Operation completed successfully\"\nfi",
"sound_alert": "# Play sound alert\nif [[ \"$OSTYPE\" == \"darwin\"* ]]; then\n afplay /System/Library/Sounds/Glass.aiff &\nelif command -v paplay &> /dev/null; then\n paplay /usr/share/sounds/freedesktop/stereo/complete.oga &\nfi"
}
}
}
},
"security_scan_code": {
"metadata": {
"name": "Security Scan Code",
"description": "Run security scans on code to detect vulnerabilities and secrets",
"use_cases": [
"Scan for secrets in code (API keys, passwords)",
"Run semgrep security rules",
"Run bandit for Python security",
"Detect common vulnerabilities"
],
"complexity": "advanced",
"safety_notes": "Can be slow (30-120 seconds). Requires security tools installed. Best for PrePush.",
"event_type": "PrePush",
"timing": "Best run before git push to catch issues early"
},
"template": {
"matcher": {},
"hooks": [
{
"type": "command",
"command": "# Run security scan\n{{SCAN_LOGIC}}",
"timeout": 120
}
]
},
"variables": {
"SCAN_TYPE": {
"description": "Type of security scan to perform",
"options": {
"secrets": "Scan for hardcoded secrets",
"semgrep": "Run semgrep security rules",
"bandit": "Run bandit for Python",
"combined": "Run all available scanners"
}
},
"SCAN_LOGIC": {
"description": "Security scanning logic",
"patterns": {
"secrets": "# Scan for secrets\necho \"Scanning for secrets...\"\nSECRETS_FOUND=false\n# Check for common secret patterns\nif git diff --cached | grep -iE '(password|api_key|secret|token)\\s*=\\s*[\"'\"'][^\"'\"']+[\"'\"']'; then\n echo \"Warning: Potential secrets found in staged changes\"\n SECRETS_FOUND=true\nfi\nif git diff --cached | grep -E '(AKIA|AWS|sk_live_|gh[ps]_)'; then\n echo \"Error: AWS/Stripe/GitHub tokens detected\"\n SECRETS_FOUND=true\nfi\nif [ \"$SECRETS_FOUND\" = true ]; then\n echo \"Security scan failed - potential secrets detected\"\n exit 1\nfi",
"semgrep": "# Run semgrep\nif ! command -v semgrep &> /dev/null; then\n echo \"semgrep not installed - skipping\"\n exit 0\nfi\necho \"Running semgrep security scan...\"\nsemgrep --config=auto --error || { echo \"Security issues found\"; exit 1; }",
"bandit": "# Run bandit for Python\nif ! command -v bandit &> /dev/null; then\n echo \"bandit not installed - skipping\"\n exit 0\nfi\necho \"Running bandit security scan...\"\nbandit -r . -f txt || { echo \"Security issues found\"; exit 1; }",
"combined": "# Run all scanners\nERRORS=0\n# Secrets scan\nif git diff --cached | grep -iE '(password|api_key|secret|token)\\s*='; then\n echo \"Warning: Potential secrets found\"\n ERRORS=$((ERRORS + 1))\nfi\n# Semgrep if available\nif command -v semgrep &> /dev/null; then\n semgrep --config=auto --error || ERRORS=$((ERRORS + 1))\nfi\n# Bandit if available\nif command -v bandit &> /dev/null; then\n bandit -r . -ll || ERRORS=$((ERRORS + 1))\nfi\nif [ $ERRORS -gt 0 ]; then\n echo \"Security scan found $ERRORS issue(s)\"\n exit 1\nfi"
}
}
}
}
}
Related skills
| Skill | Stars | Updated | Works with | Status | Install |
|---|---|---|---|---|---|
| reverse-engineer-anythingmorluto/reaReverse engineer native, managed, Electron/JavaScript, packaged, firmware, and browser targets with REA, and build agent environments (AgentEnv bundles) of web applications from a live instance or a capture, and of Linux desktop apps with a task from a request. Use shipped-artifact or requested runtime evidence to explain features, compare versions, decompile code, or guide a reconstruction. Skip REA for ordinary source-repository architecture analysis. | Stars: 471k | Updated | Claude CodeCodex | Verified | Install |
| diagnosing-superpowersobra/superpowersUse when a superpowers session went wrong and your human partner wants to know why — repeated work, ignored plans, stumbles, poor results, a skill that didn't fire, "it took too long", "why is it so expensive", "what is it doing" — or wants to build a bug report for the superpowers maintainers, for the current session or a past one identified by id or path, on any harness. | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |
| dispatching-parallel-agentsobra/superpowersUse when facing 2+ independent tasks that can be worked on without shared state or sequential dependencies | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |
| executing-plansobra/superpowersUse when executing an implementation plan in the current session as the implementer yourself — your human partner chose inline execution, or no subagent tool is available | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |