Epic Security logo

Epic Security

OrganizationPopular
epicweb-dev
epic-security

Guide on security practices including CSP, rate limiting, and session security for Epic Stack

Overview

Publisherepicweb-dev
Repositoryepic-stack
Skill nameepic-security
Stars
5.5K
Forks
461
Bundled files
Instructions only
LicenseMIT
Links
  • Markdown instructions

    A SKILL.md file the model loads on demand, so it only costs tokens when a request actually matches.

  • Works with any LLM

    AI skills are plain Markdown, not provider-specific code, so this works with GPT, Claude, Gemini, Grok, or a local model.

  • Self-contained

    Everything the model needs lives in the instructions — no extra files to sync.

  • Open source

    Published by epicweb-dev on GitHub. Read the source before you install it.

Installation

Install the Epic Security AI skill in TypingMind to use it with any LLM, or drop it into another agent that reads SKILL.md.

1

Install in TypingMind

TypingMind installs a skill straight from its GitHub folder — it reads SKILL.md, bundles the resource files, and stores the result locally.

  1. Open the app and go to Plugins → Skills.
  2. Choose "Install from GitHub".
  3. Paste the skill folder URL below and confirm.
  4. Enable the skill in any chat where you want it available.
Plugins → Skills → Add skill → From GitHub URL, then paste the folder URL and press Continue.
2

Install in another agent

Any agent that reads the Agent Skills format can use this skill — copy the folder into that agent's skills directory.

Claude Code — .claude/skills
git clone --depth 1 https://github.com/epicweb-dev/epic-stack.git /tmp/epic-stack
mkdir -p .claude/skills
cp -r /tmp/epic-stack/docs/skills/epic-security .claude/skills/epic-security
Restart Claude Code after copying so it picks up the new skill.

Use it in TypingMind

Enable Epic Security in any TypingMind chat and the model takes it from there. Its name and description sit in the system prompt, and the moment a request matches, the model loads the full instructions itself — you never invoke it by hand, and it costs no tokens until it is actually used.

The model loads Epic Security on its own as soon as a request matches it.

Works with any AI model

AI skills are plain Markdown instructions rather than provider-specific code, so Epic Security is not tied to the model it was written for. Install it once in TypingMind and use it with GPT-5, Claude, Gemini, Grok, DeepSeek, Mistral, Llama, or a local model you run yourself — all on your own API keys.

  • Loaded only when it is needed

    The system prompt carries just the name and description. The instructions are fetched on the first matching request, so an idle skill costs nothing.

  • Switch models mid-chat

    Because the skill is instructions rather than code, changing model does not break it — the next model reads the same SKILL.md.

Skill instructions

This is the SKILL.md content the model loads. Read it before installing — a skill is instructions your model will follow.

Epic Stack: Security

When to use this skill

Use this skill when you need to:

  • Configure Content Security Policy (CSP)
  • Implement spam protection (honeypot)
  • Configure rate limiting
  • Manage session security
  • Implement input validation
  • Configure secure headers
  • Manage secrets

Patterns and conventions

Security Philosophy

Following Epic Web principles:

Design to fail fast and early - Validate security constraints as early as possible. Check authentication, authorization, and input validation before processing requests. Fail immediately with clear error messages rather than allowing potentially malicious data to flow through the system.

Optimize for the debugging experience - When security checks fail, provide clear, actionable error messages that help developers understand what went wrong. Log security events with enough context to debug issues without exposing sensitive information.

Example - Fail fast validation:

typescript
// ✅ Good - Validate security constraints early
export async function action({ request }: Route.ActionArgs) {
	// 1. Authenticate immediately - fail fast if not authenticated
	const userId = await requireUserId(request)

	// 2. Validate input early - fail fast if invalid
	const formData = await request.formData()
	const submission = await parseWithZod(formData, {
		schema: NoteSchema,
	})

	if (submission.status !== 'success') {
		return data({ result: submission.reply() }, { status: 400 })
	}

	// 3. Check permissions early - fail fast if unauthorized
	await requireUserWithPermission(request, 'create:note:own')

	// Only proceed if all security checks pass
	const { title, content } = submission.value
	// ... create note
}

// ❌ Avoid - Security checks scattered or delayed
export async function action({ request }: Route.ActionArgs) {
	const formData = await request.formData()
	// ... process data first

	// Security check at the end - too late!
	const userId = await getUserId(request)
	if (!userId) {
		// Already processed potentially malicious data
		return json({ error: 'Unauthorized' }, { status: 401 })
	}
}

Example - Debugging-friendly error messages:

typescript
// ✅ Good - Clear error messages for debugging
export async function checkHoneypot(formData: FormData) {
	try {
		await honeypot.check(formData)
	} catch (error) {
		if (error instanceof SpamError) {
			// Log with context for debugging
			console.error('Honeypot triggered', {
				timestamp: new Date().toISOString(),
				userAgent: formData.get('user-agent'),
				// Don't log sensitive data
			})
			throw new Response('Form not submitted properly', { status: 400 })
		}
		throw error
	}
}

// ❌ Avoid - Generic or unhelpful errors
export async function checkHoneypot(formData: FormData) {
	try {
		await honeypot.check(formData)
	} catch (error) {
		// No context, hard to debug
		throw new Response('Error', { status: 400 })
	}
}

Content Security Policy (CSP)

Epic Stack uses CSP to prevent XSS and other attacks.

Configuration in server/index.ts:

typescript
import { helmet } from '@nichtsam/helmet/node-http'

app.use((_, res, next) => {
	helmet(res, { general: { referrerPolicy: false } })
	next()
})

Note: By default, CSP is in report-only mode to avoid blocking resources during development. In production, remove reportOnly: true to enable it fully.

Honeypot Fields

Epic Stack uses honeypot fields to protect against spam bots.

En formularios públicos:

typescript
import { HoneypotInputs } from 'remix-utils/honeypot/react'

<Form method="POST" {...getFormProps(form)}>
	<HoneypotInputs /> {/* Always include in public forms */}
	{/* Resto de campos */}
</Form>

En el action (fail fast):

typescript
import { checkHoneypot } from '#app/utils/honeypot.server.ts'

export async function action({ request }: Route.ActionArgs) {
	const formData = await request.formData()

	// Check honeypot first - fail fast if spam detected
	await checkHoneypot(formData) // Lanza error si es spam

	// Only proceed if honeypot check passes
	// ... resto del código
}

Configuration:

typescript
// app/utils/honeypot.server.ts
import { Honeypot, SpamError } from 'remix-utils/honeypot/server'

export const honeypot = new Honeypot({
	validFromFieldName: process.env.NODE_ENV === 'test' ? null : undefined,
	encryptionSeed: process.env.HONEYPOT_SECRET,
})

export async function checkHoneypot(formData: FormData) {
	try {
		await honeypot.check(formData)
	} catch (error) {
		if (error instanceof SpamError) {
			// Log for debugging (without sensitive data)
			console.error('Honeypot triggered', {
				timestamp: new Date().toISOString(),
			})
			throw new Response('Form not submitted properly', { status: 400 })
		}
		throw error
	}
}

Rate Limiting

Epic Stack uses express-rate-limit para prevenir abuso.

Basic configuration:

typescript
// server/index.ts
import rateLimit from 'express-rate-limit'

const rateLimitDefault = {
	windowMs: 60 * 1000, // 1 minute
	limit: 1000, // 1000 requests per minute
	standardHeaders: true,
	legacyHeaders: false,
	validate: { trustProxy: false },
	keyGenerator: (req: express.Request) => {
		return req.get('fly-client-ip') ?? `${req.ip}`
	},
}

const generalRateLimit = rateLimit(rateLimitDefault)

Different levels of rate limiting:

typescript
// Stricter rate limit for sensitive routes
const strongestRateLimit = rateLimit({
	...rateLimitDefault,
	limit: 10, // Only 10 requests per minute
})

// Strong rate limit for important actions
const strongRateLimit = rateLimit({
	...rateLimitDefault,
	limit: 100, // 100 requests per minute
})

Apply to specific routes:

typescript
app.use((req, res, next) => {
	const strongPaths = [
		'/login',
		'/signup',
		'/verify',
		'/admin',
		'/reset-password',
	]

	if (req.method !== 'GET' && req.method !== 'HEAD') {
		if (strongPaths.some((p) => req.path.includes(p))) {
			return strongestRateLimit(req, res, next)
		}
		return strongRateLimit(req, res, next)
	}

	return generalRateLimit(req, res, next)
})

Note: In tests and development, rate limiting is effectively disabled to allow fast tests.

Session Security

Secure session configuration:

typescript
// app/utils/session.server.ts
export const authSessionStorage = createCookieSessionStorage({
	cookie: {
		name: 'en_session',
		sameSite: 'lax', // CSRF protection advised if changing to 'none'
		path: '/',
		httpOnly: true, // Prevents access from JavaScript
		secrets: process.env.SESSION_SECRET.split(','), // Secret rotation
		secure: process.env.NODE_ENV === 'production', // HTTPS only in production
	},
})

Security features:

  • httpOnly: true - Prevents access from JavaScript (XSS protection)
  • secure: true - Only sends cookies over HTTPS in production
  • sameSite: 'lax' - CSRF protection
  • Secret rotation using array

Password Security

Hashing de passwords:

typescript
import bcrypt from 'bcryptjs'

export async function getPasswordHash(password: string) {
	const hash = await bcrypt.hash(password, 10) // 10 rounds
	return hash
}

export async function verifyUserPassword(
	where: Pick<User, 'username'> | Pick<User, 'id'>,
	password: string,
) {
	const userWithPassword = await prisma.user.findUnique({
		where,
		select: { id: true, password: { select: { hash: true } } },
	})

	if (!userWithPassword || !userWithPassword.password) {
		return null
	}

	const isValid = await bcrypt.compare(password, userWithPassword.password.hash)
	return isValid ? { id: userWithPassword.id } : null
}

Check common passwords (Have I Been Pwned):

typescript
import { checkIsCommonPassword } from '#app/utils/auth.server.ts'

const isCommonPassword = await checkIsCommonPassword(password)
if (isCommonPassword) {
	ctx.addIssue({
		path: ['password'],
		code: 'custom',
		message: 'Password is too common',
	})
}

Input Validation y Sanitization

Always validate inputs with Zod:

typescript
import { z } from 'zod'

const UserSchema = z.object({
	email: z
		.string()
		.email()
		.min(3)
		.max(100)
		.transform((val) => val.toLowerCase()),
	username: z
		.string()
		.min(3)
		.max(20)
		.regex(/^[a-zA-Z0-9_]+$/),
	password: z.string().min(6).max(72), // bcrypt limit
})

// Validate before using
const result = UserSchema.safeParse(data)
if (!result.success) {
	return json({ errors: result.error.flatten() }, { status: 400 })
}

Sanitization:

  • Use .transform() from Zod to sanitize data
  • Normalize emails to lowercase
  • Normalize usernames to lowercase
  • Clean whitespace

XSS Prevention

React prevents XSS automatically by escaping all values.

Never use dangerouslySetInnerHTML with user data:

typescript
// ❌ NEVER do this with user data
<div dangerouslySetInnerHTML={{ __html: userContent }} />

// ✅ React escapa automáticamente
<div>{userContent}</div>

Secure Headers

Epic Stack uses Helmet for secure headers.

Configuration:

typescript
import { helmet } from '@nichtsam/helmet/node-http'

app.use((_, res, next) => {
	helmet(res, { general: { referrerPolicy: false } })
	next()
})

Included headers:

  • X-Content-Type-Options: nosniff
  • X-Frame-Options: DENY
  • X-XSS-Protection: 1; mode=block
  • Referrer-Policy (configurable)

HTTPS Only

Redirect HTTP to HTTPS:

typescript
app.use((req, res, next) => {
	if (req.method !== 'GET') return next()
	const proto = req.get('X-Forwarded-Proto')
	const host = getHost(req)
	if (proto === 'http') {
		res.set('X-Forwarded-Proto', 'https')
		res.redirect(`https://${host}${req.originalUrl}`)
		return
	}
	next()
})

Secrets Management

Variables de entorno:

bash
# .env
SESSION_SECRET=secret1,secret2,secret3 # Secret rotation
HONEYPOT_SECRET=your-honeypot-secret
DATABASE_URL=file:./data/db.sqlite

En Fly.io:

bash
fly secrets set SESSION_SECRET="secret1,secret2,secret3"
fly secrets set HONEYPOT_SECRET="your-secret"

Never commit secrets:

  • Use .env.example to document required variables
  • .env is in .gitignore
  • Use fly secrets for production

Validación de Session Expiration (Fail Fast)

Always verify expiration early:

typescript
export async function getUserId(request: Request) {
	const authSession = await authSessionStorage.getSession(
		request.headers.get('cookie'),
	)
	const sessionId = authSession.get(sessionKey)

	// Fail fast - return null immediately if no session
	if (!sessionId) return null

	// Verify expiration early - fail fast if expired
	const session = await prisma.session.findUnique({
		select: { userId: true },
		where: {
			id: sessionId,
			expirationDate: { gt: new Date() }, // Verify expiration
		},
	})

	// Fail fast - destroy invalid session immediately
	if (!session?.userId) {
		throw redirect('/', {
			headers: {
				'set-cookie': await authSessionStorage.destroySession(authSession),
			},
		})
	}
	return session.userId
}

Common examples

Example 1: Public form with honeypot

typescript
// app/routes/_auth/signup.tsx
import { HoneypotInputs } from 'remix-utils/honeypot/react'
import { checkHoneypot } from '#app/utils/honeypot.server.ts'

export async function action({ request }: Route.ActionArgs) {
	const formData = await request.formData()

	await checkHoneypot(formData) // Spam protection

	const submission = await parseWithZod(formData, {
		schema: SignupSchema,
	})

	// ... rest of code
}

export default function SignupRoute({ actionData }: Route.ComponentProps) {
	return (
		<Form method="POST" {...getFormProps(form)}>
			<HoneypotInputs /> {/* Include in form */}
			{/* Rest of fields */}
		</Form>
	)
}

Example 2: Custom rate limiting

typescript
// server/index.ts
const apiRateLimit = rateLimit({
	...rateLimitDefault,
	windowMs: 60 * 1000,
	limit: 100, // 100 requests per minute for API
	keyGenerator: (req) => {
		const apiKey = req.get('X-API-Key')
		return apiKey ?? req.get('fly-client-ip') ?? req.ip
	},
})

app.use('/api', apiRateLimit)

Example 3: Strict input validation

typescript
// app/utils/user-validation.ts
import { z } from 'zod'

export const EmailSchema = z
	.string({ required_error: 'Email is required' })
	.email({ message: 'Email is invalid' })
	.min(3, { message: 'Email is too short' })
	.max(100, { message: 'Email is too long' })
	.transform((value) => value.toLowerCase().trim()) // Sanitization

export const UsernameSchema = z
	.string({ required_error: 'Username is required' })
	.min(3, { message: 'Username is too short' })
	.max(20, { message: 'Username is too long' })
	.regex(/^[a-zA-Z0-9_]+$/, {
		message: 'Username can only include letters, numbers, and underscores',
	})
	.transform((value) => value.toLowerCase().trim()) // Sanitization

export const PasswordSchema = z
	.string({ required_error: 'Password is required' })
	.min(6, { message: 'Password is too short' })
	.refine((val) => new TextEncoder().encode(val).length <= 72, {
		message: 'Password is too long', // bcrypt limit
	})

Example 4: Permission verification before actions

typescript
export async function action({ request }: Route.ActionArgs) {
	const userId = await requireUserId(request)

	// Validate that user has permission
	await requireUserWithPermission(request, 'delete:note:own')

	// Only after validating permissions
	await prisma.note.delete({ where: { id: noteId } })

	return redirect('/notes')
}

Common mistakes to avoid

  • Delayed security checks: Validate authentication, authorization, and input as early as possible - fail fast
  • Generic error messages: Provide clear, actionable error messages that help with debugging (without exposing sensitive data)
  • Forgetting honeypot in public forms: Always include HoneypotInputs in forms accessible without authentication
  • Not validating session expiration: Always verify expirationDate when getting sessions - check early
  • Using dangerouslySetInnerHTML with user data: Never render user HTML without sanitizing
  • Not using rate limiting: Protect sensitive routes with rate limiting
  • Secrets in code: Never hardcode secrets, use environment variables
  • Not sanitizing inputs: Always sanitize inputs with .transform() from Zod
  • Not validating common passwords: Check passwords against Have I Been Pwned
  • Sessions without httpOnly: Always use httpOnly: true in session cookies
  • Not using HTTPS in production: Make sure to redirect HTTP to HTTPS
  • CSP too permissive: Review and adjust CSP according to your needs
  • Not logging security events: Log security failures with context for debugging (without sensitive data)

References

Frequently asked questions

What does the Epic Security AI skill do?

Guide on security practices including CSP, rate limiting, and session security for Epic Stack

Why use Epic Security on TypingMind?

Because you install it once and use it with any model. Epic Security is plain Markdown rather than provider-specific code, so the same skill runs on GPT-5, Claude, Gemini, Grok, or a local model — and you can switch model mid-chat without it breaking. TypingMind runs on your own API keys, so you pay providers directly instead of a per-seat subscription, and your skills and chats stay in your own storage.

How do I install Epic Security in TypingMind?

Open Plugins → Skills → Install from GitHub in TypingMind and paste https://github.com/epicweb-dev/epic-stack/tree/main/docs/skills/epic-security. TypingMind reads its SKILL.md and installs it as a skill you can enable per chat.

Which AI models can use Epic Security?

Any model you connect in TypingMind. AI skills are plain Markdown instructions rather than provider-specific code, so GPT, Claude, Gemini, Grok, and local models can all load this skill when a request matches it.

How many AI models can I use with Epic Security?

As many as you like. As long as a model supports skills, you can use Epic Security with it — GPT, Claude, Gemini, Grok, DeepSeek, Mistral, Llama and more — all on TypingMind with your own API keys.

Is the Epic Security AI skill free?

Yes. It is published on GitHub by epicweb-dev under the MIT license. You only pay your own AI provider for the tokens you use.

What are AI skills?

An AI skill is a reusable instruction bundle that teaches an AI model how to do one specific task. It follows the open Agent Skills format: a SKILL.md file with a name and description, plus any scripts, templates or reference files the model may need. The model reads the instructions only when your request matches the skill, so an installed skill costs nothing until it is used.

How are AI skills different from plugins or MCP servers?

A plugin or MCP server gives a model new tools to call — code that runs somewhere and returns a result. An AI skill gives the model knowledge and process instead: how to approach a task, which steps to follow, what good output looks like. Skills are plain Markdown, so they need no server, no API key and no runtime, and they work with any model.

View all

Set up your own AI workspace now

Get notified about new features and future giveaways by subscribing to our newsletter 👇