The Problem
AI coding assistants generate functional code fast — but they routinely omit security controls. A request for "a login endpoint" produces working code that skips password hashing, has no rate limiting, stores sessions in memory, and concatenates SQL strings. The code compiles. The tests pass. The vulnerabilities ship to production.
This happens because LLMs optimize for the task described, not the security context implied. Without explicit instructions, they take the shortest path to functional code.
How It Works
Manicode prompts are system instructions — not plugins, agents, or wrappers. You set a prompt as the system message, and the LLM's output changes.
Pick your technology
Find the prompt for your stack and version in the registry — Express 5.2, Django 6.0, Spring Boot 4.0, or any of 300+ technologies.
Drop it in
Set the prompt as the system instruction in your IDE (Cursor, Copilot, Devin Desktop), coding agent (Claude Code, Codex), web chat (ChatGPT, Claude), or API call. One file, no dependencies.
Get secure code
Ask for code the way you normally would. The LLM generates functional code that includes security controls by default — no extra prompting required.
Before and After
Same model, same request, same framework. The only difference is whether a Manicode prompt is active as the system instruction.
// "Write a login endpoint"
app.post('/login', async (req, res) => {
const { email, password } = req.body;
const user = await db.query(
`SELECT * FROM users WHERE email = '${email}'`
);
if (user && password === user.password) {
req.session.user = user;
return res.json({ success: true });
}
res.status(401).json({ error: 'Invalid' });
});
- SQL injection via string interpolation
- Plaintext password comparison
- No rate limiting on auth endpoint
- No input validation
- Session fixation (no regeneration)
// Same request, with the Manicode Express 5.2 prompt active
app.post('/login',
rateLimit({ windowMs: 15 * 60 * 1000, limit: 5 }),
validate(loginSchema),
async (req, res) => {
const { email, password } = req.validated.body;
const user = await db.query(
'SELECT id, email, password_hash FROM users WHERE email = $1',
[email]
);
if (!user || !(await argon2.verify(user.password_hash, password))) {
return res.status(401).json({ error: 'Invalid credentials' });
}
await new Promise((resolve, reject) =>
req.session.regenerate((err) => (err ? reject(err) : resolve())));
req.session.userId = user.id;
res.json({ success: true });
});
- Parameterized query prevents SQL injection
- Argon2 password hashing
- Rate limiting (5 attempts / 15 min)
- Schema validation on input
- Session regeneration after login
What You Get
Standards Coverage
- Official sources — Each prompt is written from the technology's official documentation and security advisories for that exact version
- CWE — Every prompt's rules are mapped to the Common Weakness Enumeration weaknesses they defend against. See the mappings in the registry.
- Required topics — Every web framework prompt covers ten core topics, from password storage and CSRF to SSRF and file uploads
- OWASP Top 10 for LLM Applications — AI and MCP prompts cover prompt injection, tool authorization, output handling, and the other LLM application risks it describes
Versioned Prompts
- One prompt per version — Django 5.2, 6.0, and 6.1 each get their own prompt, written for the APIs and defaults of that release
- Works with any model — The same prompt works with Claude, GPT, Gemini, and any other model that accepts a system instruction
- Reviewed dates — Every prompt records when it was last reviewed
Deploy Anywhere
Manicode prompts are plain markdown files. They work with any tool that accepts a system instruction — no SDK, no plugin, no vendor lock-in.
IDEs
- GitHub Copilot
- Cursor
- Devin Desktop
- JetBrains AI
- Kiro
- Cline
Coding Agents
- Claude Code
- OpenAI Codex
- Aider
- Cursor Agent
- Kiro CLI
Web Chat
- ChatGPT Projects
- Claude Projects
- Google AI Studio
APIs
- Anthropic API
- OpenAI API
- Google Gemini API
- OpenRouter
Inside a Prompt
A prompt includes:
- Base rules — The rules for the language and runtime the technology builds on, such as JavaScript and Node.js for Express
- Technology rules — Rules grouped by topic for that technology's APIs: request parsing, sessions, authorization, responses, deployment, and more
Prompts are dense, technical, and opinionated — because vague security guidance produces vague security controls. See what's in a prompt.
About This Portal
This documentation portal is your guide to Manicode's code security prompt library. Here you will find:
- Prompt library — What a prompt contains and how prompts are named and versioned
- Deployment guides — Step-by-step instructions for every major IDE, coding agent, web chat interface, and API
- Prompt delivery — How prompt updates reach your prompt repo and how to keep your copies current
- Security assumptions — What prompts do and don't guarantee, and where trust boundaries lie
Start Generating Secure Code
Copy your prompt from your prompt repo into your tool, and ask for code the way you normally would.