We take the security of Codesi Programming Language seriously. This document outlines our security policies and how to report vulnerabilities.

Note: We recommend always using the latest stable version for the best security and features.

Codesi has been designed with security in mind:

1. No External Dependencies: Zero third-party libraries reduce attack surface
2. Safe Evaluation: No use of Python's dangerous eval() function
3. Scope Isolation: Proper variable scoping prevents unintended access
4. Input Sanitization: File operations validate paths
5. Error Handling: Comprehensive exception handling prevents crashes
6. Type Safety: Strong type checking at runtime

# Safe file operations
- Path validation
- Permission checking
- Error handling for all file I/O
- No arbitrary code execution

# Codesi interpreter
- AST-based interpretation (no eval)
- Sandboxed scope execution
- Controlled function calls
- Safe built-in functions only

DO NOT create a public GitHub issue for security vulnerabilities.

Instead, report vulnerabilities privately to:

Email: codesilang@gmail.com GitHub: Send a private security advisory at https://github.com/codesi-lang
Direct Contact: @rishaankgupta

A good security report should include:

## Vulnerability Type
[e.g., Code Injection, Path Traversal, DoS, etc.]

## Affected Version
Codesi version number (e.g., 0.0.1)

## Description
Clear description of the vulnerability

## Steps to Reproduce
1. Create file with: ...
2. Run command: ...
3. Observe: ...

## Impact
What can an attacker do with this?

## Proof of Concept
```codesi
// Minimal code demonstrating the issue

If you have ideas for fixing it

• OS: Windows/macOS/Linux
• Python version: 3.x
• Additional context

### Response Timeline

We aim to respond to security reports within:

- **Acknowledgment**: 48 hours
- **Initial Assessment**: 5 business days
- **Fix Development**: Depends on severity
  - **Critical**: 7 days
  - **High**: 14 days
  - **Medium**: 30 days
  - **Low**: Next release
- **Public Disclosure**: After fix is released

### Disclosure Policy

We follow responsible disclosure:

1. **Private Report**: You report the issue privately
2. **Acknowledgment**: We acknowledge receipt
3. **Investigation**: We investigate and develop a fix
4. **Testing**: We test the fix thoroughly
5. **Release**: We release a patched version
6. **Public Disclosure**: We publicly disclose after users have time to update (typically 7-14 days)
7. **Credit**: We credit you (unless you prefer anonymity)

---

## 🎯 Security Vulnerability Categories

### Critical 🔴

Vulnerabilities that:
- Allow arbitrary code execution
- Enable privilege escalation
- Expose sensitive data without authentication
- Cause complete system compromise

**Response Time**: Immediate (within 7 days)

### High 🟠

Vulnerabilities that:
- Allow unauthorized data access
- Enable authentication bypass
- Cause significant data loss
- Lead to denial of service

**Response Time**: 14 days

### Medium 🟡

Vulnerabilities that:
- Expose limited information
- Require user interaction
- Have limited impact scope
- Affect only specific configurations

**Response Time**: 30 days

### Low 🟢

Vulnerabilities that:
- Require complex attack chains
- Have minimal real-world impact
- Affect only edge cases
- Are easily mitigated by users

**Response Time**: Next release

---

## 🔐 Known Security Considerations

### Current Design Decisions

1. **File System Access**
   - Codesi can read/write files with user permissions
   - Users should be careful with untrusted `.cds` files
   - No privilege escalation possible

2. **REPL Mode**
   - Full access to interpreter features
   - Runs with current user permissions
   - Use caution when running untrusted code

3. **Import System**
   - Can import other Codesi files
   - Imported files execute with same permissions
   - Be cautious with untrusted imports

### Best Practices for Users

```codesi
// ✅ Safe: Read from controlled locations
content = file_padho("data/config.txt")

// ⚠️  Caution: User-provided paths
user_path = input_lo("File path: ")
// Validate before using!

// ❌ Risky: Blindly trusting user input
// Always validate and sanitize!

1. Validate Input

   // Always validate user input
   naam = input_lo("Naam: ")
   agar (naam.lambai() > 100) {
       throw {message: "Naam bahut lamba hai"}
   }
   

2. Check File Paths

   // Validate file paths before using
   karya safe_read(path) {
       // Check if file exists
       agar (!file_hai(path)) {
           throw {message: "File nahi mili"}
       }
       vapas file_padho(path)
   }
   

3. Handle Errors

   // Use try-catch for file operations
   try {
       content = file_padho("data.txt")
   } catch (e) {
       likho("Error:", e.message)
   }
   

1. Run with Minimal Permissions: Don't run Codesi as root/admin
2. Isolated Directories: Use separate directories for untrusted scripts
3. Virtual Environments: Use Python virtual environments
4. Code Review: Review .cds files before running
5. Backup Data: Regular backups before running new scripts

• Date: 2025-10-XX
• Version: 0.0.1
• Reviewer: Rishaank Gupta
• Status: No critical issues found

• ✅ Lexer and tokenizer
• ✅ Parser and AST generation
• ✅ Interpreter execution
• ✅ Built-in functions
• ✅ File I/O operations
• ✅ Scope management
• ✅ Error handling

• Input fuzzing
• Path traversal attempts
• Code injection attempts
• Memory leak testing
• Error handling validation

We appreciate security researchers who help keep Codesi secure.

Security researchers who responsibly disclose vulnerabilities will be:

• Credited in the security advisory (unless they prefer anonymity)
• Listed in our Hall of Fame
• Mentioned in release notes
• Given our sincere thanks!

No reported vulnerabilities yet - be the first!

• OWASP Top 10
• Python Security
• Secure Coding Guidelines

• Contributing Guidelines
• Code of Conduct
• Issue Tracker

This security policy may be updated from time to time. Major changes will be announced through:

• GitHub releases
• Repository README
• Security advisories

Last Updated: 2025-10-XX
Version: 0.0.1

Before submitting code that touches security-sensitive areas:

• No use of Python eval() or exec()
• Input validation for all user-provided data
• Path validation for file operations
• Proper error handling with no sensitive info leaks
• No hardcoded credentials or secrets
• Safe defaults for all options
• Documentation of security implications
• No introduction of new dependencies
• Testing for edge cases and attacks

For critical, actively exploited vulnerabilities:

Contact: codesilang@gmail.com (Priority response)
GitHub Security Advisory: https://github.com/codesi-lang Expected Response: Within 24 hours

• Status: No known vulnerabilities
• Release Date: 2025-10-XX
• Security Improvements: Initial secure implementation

Future vulnerabilities and fixes will be documented here

We believe in:

• Transparency: Open communication about security
• Collaboration: Working together with researchers
• Respect: Treating all reporters with respect
• Responsibility: Quick action on valid reports
• Recognition: Crediting those who help us

Security Team: codesilang@gmail.com
Project Lead: Rishaank Gupta (@rishaankgupta)
GitHub: https://github.com/codesi-lang