MyAppMyWeb is committed to maintaining the highest security posture for autonomous agent workflows and untrusted code execution.

MyAppMyWeb processes natural language requirements to generate and execute full-stack code. To prevent Arbitrary Code Execution (ACE) and Remote Code Execution (RCE) vulnerabilities:

1. Ephemeral Sandboxing: All user-generated code is executed inside ephemeral, isolated Docker containers or isolated sandbox processes with read-only root filesystems.
2. Network Egress Isolation: Sandbox containers operate with --network none by default to prevent unauthorized telemetry, reverse shells, or data exfiltration.
3. Resource & PID Throttling: Hard memory ceilings (512MB), CPU quota caps (1.0 Core), and PID limits (max 64 processes) prevent fork-bombs and runaway resource exhaustion.
4. Non-Root Execution: Sandboxed tasks run under dedicated unprivileged system users (UID 1000:1000) with all Linux capabilities dropped (--cap-drop=ALL).
5. AST & Static Pre-Screening: Code is statically audited for hazardous system primitives (child_process, os.system, /etc/passwd reads) prior to execution.

If you discover a potential security vulnerability in MyAppMyWeb:

1. Do NOT open a public issue.
2. Send a detailed report directly to the security maintainers via email at ishant.security@gmail.com or submit an advisory through GitHub Private Vulnerability Reporting.
3. Include:
   • Reproduction steps or proof-of-concept (PoC) prompt.
   • Affected agent component (Manager, Developer, Tester, Sandbox).
   • Potential impact.

We follow standard Responsible Disclosure timelines and aim to acknowledge reports within 48 hours and release patches within 14 days.