A secure, server-side HTTP client with built-in API key validation, rate limiting, and security features.
- ๐ Server-side only execution
- ๐ API key validation and requirement
- โฑ๏ธ Rate limiting with Redis
- ๐ HTTPS enforcement
- โก Request timeout handling
- ๐ก๏ธ Comprehensive error handling
npm install secure-server-fetch
# or
yarn add secure-server-fetch
Create a .env
file in your project root:
UPSTASH_REDIS_REST_URL=your_redis_url
UPSTASH_REDIS_REST_TOKEN=your_redis_token
Both variables are required for rate limiting functionality. The Redis URL must be a valid HTTPS URL.
import { secureServerFetch } from 'secure-server-fetch';
async function fetchData() {
try {
const data = await secureServerFetch('https://api.example.com/data', {
apiKey: 'your-api-key',
timeout: 5000, // 5 seconds
});
return data;
} catch (error) {
console.error('Fetch failed:', error);
}
}
import { requireApiKey } from 'secure-server-fetch';
// In your API route handler
export async function handler(request: Request) {
const apiKeyCheck = requireApiKey(request, 'your-expected-api-key');
if (apiKeyCheck) {
return apiKeyCheck; // Returns 401 response if validation fails
}
// Continue with your API logic
}
The API key validation enforces the following requirements:
- Length: Must be at least 32 characters long
- Character Set: Can only contain:
- Uppercase letters (A-Z)
- Lowercase letters (a-z)
- Numbers (0-9)
- Underscores (_)
- Hyphens (-)
- Character Mix: Must contain at least:
- One uppercase letter
- One lowercase letter
- One number
All error responses are returned in JSON format with appropriate HTTP status codes:
- Missing API Key (Status: 401)
{
"error": "API key is missing",
"message": "Please provide 'x-api-key' header with a valid API key",
"requirements": {
"format": "Must be at least 32 characters long",
"characters": "Can only contain letters, numbers, underscores, and hyphens",
"complexity": "Must contain at least one uppercase letter, one lowercase letter, and one number"
}
}
- Empty API Key (Status: 401)
{
"error": "Empty API key",
"message": "API key cannot be empty",
"requirements": {
"format": "Must be at least 32 characters long",
"characters": "Can only contain letters, numbers, underscores, and hyphens",
"complexity": "Must contain at least one uppercase letter, one lowercase letter, and one number"
}
}
- Invalid Format (Status: 401)
{
"error": "Invalid API key format",
"message": "[Specific validation error message]",
"requirements": {
"format": "Must be at least 32 characters long",
"characters": "Can only contain letters, numbers, underscores, and hyphens",
"complexity": "Must contain at least one uppercase letter, one lowercase letter, and one number"
}
}
- Invalid API Key (Status: 401)
{
"error": "Invalid API key",
"message": "The provided API key is not valid"
}
import { rateLimit } from 'secure-server-fetch';
async function handleRequest(request: Request) {
try {
const rateLimitResult = await rateLimit(request, 'unique-identifier', {
maxRequests: 100,
timeframeMs: 60000, // 1 minute
});
// Rate limit headers will be automatically included in rateLimitResult.headers
// Your API logic here
} catch (error) {
if (error.name === 'RateLimitError') {
return new Response('Rate limit exceeded', { status: 429 });
}
}
}
Makes a secure server-side HTTP request.
apiKey?: string
- API key for authenticationrequireHttps?: boolean
- Enforce HTTPS (default: true)timeout?: number
- Request timeout in ms (default: 30000)...RequestInit
- All standard fetch options
Validates API key from request headers.
request: Request
- Incoming request objectexpectedKey: string
- The API key to validate against
Implements rate limiting for requests.
maxRequests: number
- Maximum requests allowedtimeframeMs: number
- Time window in millisecondsprefix?: string
- Redis key prefix
-
API Keys
- Use keys at least 32 characters long
- Include mix of uppercase, lowercase, and numbers
- Rotate keys periodically
- Store keys securely in environment variables
-
Rate Limiting
- Implement rate limiting on all public endpoints
- Use unique identifiers (e.g., IP, user ID)
- Set appropriate limits based on endpoint sensitivity
-
HTTPS
- Always use HTTPS in production
- Only disable HTTPS requirement in development
-
Error Handling
- Never expose internal errors to clients
- Use provided error classes for consistent handling
- Log errors securely
The library provides specific error classes:
ServerSideError
: For client-side execution attemptsNetworkError
: For network and HTTP errorsValidationError
: For input validation failuresRateLimitError
: For rate limit violations
Contributions are welcome! Please read our contributing guidelines and submit pull requests.
MIT