Skip to content

Security: AKIVA-AI/toolkit-data-contracts

Security

SECURITY.md

Security Policy

Supported Versions

Version Supported
0.1.x Yes

Reporting a Vulnerability

Do NOT open a public GitHub issue for security vulnerabilities.

Email security@akiva.com with:

  • Description of the vulnerability
  • Steps to reproduce
  • Impact assessment
  • Affected version(s)

We will acknowledge receipt within 48 hours and provide a triage assessment within 7 business days.

Disclosure Policy

We follow coordinated disclosure. We ask that you give us 90 days to address the issue before public disclosure.

Scope

In Scope

  • Code in the src/ directory
  • CLI argument parsing and file handling
  • Contract inference, validation, and drift detection logic
  • Docker image configuration
  • CI/CD pipeline security (workflow injection, secret exposure)

Out of Scope

  • Issues in upstream dependencies (report to the upstream project)
  • Issues requiring physical access to the machine running the tool
  • Social engineering attacks

Security Considerations

This tool processes local files only. It has zero runtime dependencies and does not make network connections. Key security properties:

  • No eval/exec: No dynamic code execution of any kind
  • No shell commands: No subprocess or os.system calls
  • No network access: Purely file-based I/O
  • Path validation: All file paths are resolved to absolute paths before access
  • Non-root Docker: Container runs as unprivileged user

There aren't any published security advisories