env-manager
Environment Variable Manager (env-manager)
Comprehensive environment variable validation, security scanning, and management for modern web applications.
Overview
The env-manager skill provides systematic environment variable management across local development, CI/CD pipelines, and deployment platforms. It prevents common issues like missing variables, exposed secrets, and framework-specific configuration errors.
Key Features:
- Framework-Aware Validation: Next.js, Vite, React, Node.js, Flask support
- Security-First: Never logs secrets, detects exposed credentials
- Platform Integration: Ready for Vercel, Railway, Heroku, and CI/CD
- Fast: Validates 1000 variables in 0.025s (80x faster than 2s target)
- Zero Dependencies: Pure Python, works anywhere
Why Use env-manager?
Common problems this solves:
- "Works on my machine, but not in production" (missing env vars)
- Accidentally exposing API keys in client-side code (NEXT_PUBLIC_ with secrets)
- Missing required variables during deployment
- Inconsistent .env files across team members
- No documentation of required environment variables
- Security vulnerabilities from exposed secrets
Quick Start
Installation
No installation needed! env-manager is a bundled skill in Claude MPM.
Requirements:
- Python 3.7+
- No external dependencies
5-Minute Quick Start
# 1. Validate your .env file
python3 scripts/validate_env.py .env
# 2. Check for framework-specific issues (Next.js example)
python3 scripts/validate_env.py .env --framework nextjs
# 3. Compare with .env.example to find missing vars
python3 scripts/validate_env.py .env --compare-with .env.example
# 4. Generate .env.example for documentation
python3 scripts/validate_env.py .env --generate-example .env.example
# 5. Get JSON output for CI/CD integration
python3 scripts/validate_env.py .env --json
That's it! Environment variables are now validated professionally.
Usage Examples
Basic Validation
Validate a .env file for structural issues:
python3 scripts/validate_env.py .env
What it checks:
- Valid key=value format
- No duplicate keys
- Proper naming conventions (UPPERCASE_WITH_UNDERSCORES)
- No empty values (unless explicitly allowed)
- Proper quoting for values with spaces
Example output:
✅ Validation successful!
- 15 variables validated
- 0 errors
- 0 warnings
Framework-Specific Validation
Next.js
Validate Next.js environment variables:
python3 scripts/validate_env.py .env.local --framework nextjs
Next.js-specific checks:
- NEXT_PUBLIC_* variables are client-exposed (warns if secrets detected)
- .env.local, .env.production, .env file hierarchy
- Detects secrets in client-side variables
Example:
# .env.local
NEXT_PUBLIC_API_URL=https://api.example.com
NEXT_PUBLIC_API_KEY=secret123 # ⚠️ WARNING: Secret in client-exposed variable!
DATABASE_URL=postgresql://... # ✅ Server-side only
Vite
python3 scripts/validate_env.py .env --framework vite
Vite-specific checks:
- VITE_* variables are client-exposed
- Warns if secrets detected in VITE_ prefixed vars
React (Create React App)
python3 scripts/validate_env.py .env --framework react
React-specific checks:
- REACT_APP_* variables are client-exposed
- Warns if secrets in REACT_APP_ prefixed vars
Node.js/Express
python3 scripts/validate_env.py .env --framework nodejs
Node.js-specific checks:
- Common NODE_ENV, PORT, DATABASE_URL patterns
- Standard Node.js conventions
Flask/Python
python3 scripts/validate_env.py .env --framework flask
Flask-specific checks:
- FLASK_APP, FLASK_ENV variables
- SQLAlchemy DATABASE_URL format
Comparing with .env.example
Ensure your .env has all required variables:
python3 scripts/validate_env.py .env --compare-with .env.example
What it checks:
- All variables in .env.example exist in .env
- No extra undocumented variables in .env
Example output:
❌ Missing variables:
- DATABASE_URL (required in .env.example)
- STRIPE_SECRET_KEY (required in .env.example)
⚠️ Extra variables not in .env.example:
- DEBUG_MODE (consider adding to .env.example)
Perfect for:
- Onboarding new team members
- CI/CD validation
- Deployment pre-checks
Generating .env.example
Create documentation for your environment variables:
python3 scripts/validate_env.py .env --generate-example .env.example
What it does:
- Reads your .env file
- Sanitizes secret values (replaces with placeholders)
- Generates .env.example with safe defaults
Example:
# Input: .env
DATABASE_URL=postgresql://user:pass@localhost/db # pragma: allowlist secret
STRIPE_SECRET_KEY=sk_live_abc123xyz
NEXT_PUBLIC_API_URL=https://api.example.com
# Output: .env.example
DATABASE_URL=postgresql://user:password@localhost/dbname # pragma: allowlist secret
STRIPE_SECRET_KEY=your_stripe_secret_key_here
NEXT_PUBLIC_API_URL=https://api.example.com
Security note: env-manager detects common secret patterns and replaces them with safe placeholders.
CI/CD Integration
Get machine-readable JSON output for automated workflows:
python3 scripts/validate_env.py .env.example --strict --json
JSON output format:
{
"valid": true,
"errors": [],
"warnings": [],
"stats": {
"total_vars": 15,
"errors": 0,
"warnings": 0
}
}
Exit codes:
0: Validation passed1: Validation errors found2: Missing required file3: Warnings found (only in --strict mode)
GitHub Actions example:
name: Validate Environment Variables
on: [push, pull_request]
jobs:
validate-env:
runs-on: ubuntu-latest
steps:
- uses: actions/checkout@v3
- name: Validate .env.example
run: |
python3 scripts/validate_env.py .env.example --strict --json
working-directory: ./path/to/skill
- name: Check for framework-specific issues
run: |
python3 scripts/validate_env.py .env.example --framework nextjs --json
working-directory: ./path/to/skill
Strict Mode
Treat warnings as errors (useful for CI/CD):
python3 scripts/validate_env.py .env --strict
When to use:
- Pre-deployment validation
- CI/CD pipelines
- Release gates
- Team standard enforcement
Quiet Mode
Show only errors, suppress warnings:
python3 scripts/validate_env.py .env --quiet
When to use:
- You've already reviewed warnings
- Automated scripts that only care about errors
- Noisy environments where warnings are distracting
Supported Frameworks
| Framework | Prefix | Client-Exposed | Notes |
|---|---|---|---|
| Next.js | NEXT_PUBLIC_* |
Yes | Auto-exposed in browser |
| Vite | VITE_* |
Yes | Bundled into client code |
| React (CRA) | REACT_APP_* |
Yes | Embedded in production build |
| Node.js | N/A | No | Server-side only |
| Flask | N/A | No | Server-side only |
Security warning: Never put secrets in client-exposed variables (NEXT_PUBLIC_, VITE_, REACT_APP_). env-manager will warn you if it detects common secret patterns.
CLI Reference
Command Structure
python3 scripts/validate_env.py <file> [options]
Options
| Option | Description | Example |
|---|---|---|
--compare-with FILE |
Compare with .env.example | --compare-with .env.example |
--framework {nextjs|vite|react|nodejs|flask|generic} |
Framework-specific validation | --framework nextjs |
--strict |
Treat warnings as errors | --strict |
--json |
JSON output for automation | --json |
--quiet |
Only show errors | --quiet |
--generate-example OUTPUT |
Generate .env.example | --generate-example .env.example |
Exit Codes
| Code | Meaning | When |
|---|---|---|
0 |
Success | No errors (warnings OK unless --strict) |
1 |
Validation errors | Structural issues, duplicates, etc. |
2 |
File not found | Specified .env file doesn't exist |
3 |
Warnings in strict mode | Warnings exist and --strict enabled |
Common Use Cases
Scenario 1: New Developer Onboarding
# New developer clones repo
git clone <repo>
cd <project>
# Copy example and fill in values
cp .env.example .env
# Edit .env with actual values...
# Validate setup
python3 scripts/validate_env.py .env --compare-with .env.example
# If missing variables, fix them
# Validation passes ✅
Scenario 2: Pre-Deployment Check
# Before deploying to Vercel/Railway/Heroku
python3 scripts/validate_env.py .env.production --framework nextjs --strict
# Fix any errors
# Deploy with confidence ✅
Scenario 3: Security Audit
# Check for accidentally exposed secrets
python3 scripts/validate_env.py .env.local --framework nextjs
# Look for warnings like:
# ⚠️ NEXT_PUBLIC_STRIPE_SECRET: Contains potential secret in client-exposed variable
Scenario 4: Team Documentation
# After adding new environment variable
echo "NEW_API_KEY=abc123" >> .env
# Regenerate .env.example
python3 scripts/validate_env.py .env --generate-example .env.example
# Commit updated .env.example
git add .env.example
git commit -m "docs: add NEW_API_KEY to environment variables"
Scenario 5: CI/CD Quality Gate
# In your CI pipeline
- name: Validate environment configuration
run: |
python3 scripts/validate_env.py .env.example --strict --json > validation.json
# Fail pipeline if validation fails
if [ $? -ne 0 ]; then
cat validation.json
exit 1
fi
Performance
env-manager is designed for speed:
Benchmarks:
- Validates 1000 variables in 0.025s
- 80x faster than 2s target
- Zero external dependencies
- Minimal memory footprint
Why it matters:
- Fast feedback during development
- No CI/CD slowdown
- Works in resource-constrained environments
Security Notes
Critical security features:
- Never Logs Secrets: env-manager NEVER displays actual secret values in output
- Client-Exposure Detection: Warns when secrets are in NEXT_PUBLIC_, VITE_, REACT_APP_ variables
- Secret Sanitization: When generating .env.example, replaces secrets with safe placeholders
- No Network Calls: All validation is local, no data leaves your machine
Security-audited: This skill has undergone security review. See references/security.md for details.
Best practices:
- Never commit .env files with secrets
- Always use .env.example for documentation
- Use platform secret managers (Vercel, Railway, etc.) for production
- Validate before every deployment
- Run security scan regularly
Common Issues
"Missing equals sign" error
Cause: Line in .env doesn't have = separator
Fix:
# ❌ Bad
API_KEY
# ✅ Good
API_KEY=your_key_here
"Duplicate key" error
Cause: Same variable defined multiple times
Fix:
# ❌ Bad
API_KEY=value1
API_KEY=value2
# ✅ Good
API_KEY=value2
"Invalid variable name" warning
Cause: Variable name doesn't follow UPPERCASE_WITH_UNDERSCORES convention
Fix:
# ❌ Bad
apiKey=value
api-key=value
# ✅ Good
API_KEY=value
"Potential secret in client-exposed variable" warning
Cause: NEXT_PUBLIC_, VITE_, or REACT_APP_ variable contains secret-like value
Fix:
# ❌ Bad (secret exposed to client!)
NEXT_PUBLIC_STRIPE_SECRET=sk_live_abc123
# ✅ Good (server-side only)
STRIPE_SECRET_KEY=sk_live_abc123
NEXT_PUBLIC_STRIPE_PUBLISHABLE=pk_live_xyz789
"Empty value" warning
Cause: Variable has no value
Fix:
# ❌ Bad
DATABASE_URL=
# ✅ Good (if optional, document it)
DATABASE_URL= # Optional, uses SQLite if not set
# ✅ Better
DATABASE_URL=postgresql://localhost/mydb
File not found error
Cause: Specified .env file doesn't exist
Fix:
# Check file exists
ls -la .env
# Or create it
touch .env
Troubleshooting
Validation passes locally but fails in CI
Check:
- Line endings (CRLF vs LF)
- File encoding (UTF-8 expected)
- File permissions
- Python version (3.7+ required)
Warnings about client-exposed variables
This is intentional! env-manager is warning you that variables like NEXT_PUBLIC_API_KEY will be visible in the browser.
Options:
- Move secret to server-side variable (remove NEXT_PUBLIC_ prefix)
- Use public/publishable keys only in client-exposed variables
- If truly not a secret, ignore the warning
.env.example generation replaces too much
env-manager is conservative about secrets. If it over-sanitizes:
- Manually edit .env.example after generation
- Use specific placeholder values in .env that won't trigger sanitization
Advanced Usage
Custom Validation Patterns
See references/validation.md for advanced validation patterns.
Platform-Specific Deployment
See references/synchronization.md for Vercel, Railway, Heroku integration patterns.
Framework-Specific Guides
See references/frameworks.md for comprehensive framework guides.
Related Documentation
- Validation Reference: Complete validation workflows and checks
- Security Reference: Secret scanning and security patterns
- Synchronization Reference: Platform sync patterns (coming soon)
- Frameworks Reference: Framework-specific patterns and conventions
- Troubleshooting Guide: Common issues and solutions
Integration with Claude MPM
env-manager is a bundled skill in Claude MPM. Agents can use it for:
- Pre-deployment validation
- Security scanning
- Environment setup verification
- Documentation generation
See INTEGRATION.md for agent integration patterns.
Contributing
env-manager follows Claude MPM contribution guidelines:
- Run
make lint-fixduring development - Run
make qualitybefore commits - Add tests for new features (85%+ coverage required)
- Update documentation
See CONTRIBUTING.md for details.
License
MIT License - Part of Claude MPM project
Support
- Issues: Report bugs via GitHub Issues
- Documentation: See references/ directory
- Examples: See examples/ directory
- Integration: See INTEGRATION.md
Version: 1.0.0 Status: Stable, Security-Audited Test Coverage: 85%+ Performance: 80x faster than target