Skip to content

Security: vipasane/agentscope

Security

SECURITY.md

Security Policy

Supported Versions

Version Supported
1.x.x
< 1.0

Reporting a Vulnerability

We take security seriously. If you discover a security vulnerability, please report it responsibly.

How to Report

  1. Do NOT open a public GitHub issue for security vulnerabilities
  2. Email the maintainers directly or use GitHub's private vulnerability reporting
  3. Include:
    • Description of the vulnerability
    • Steps to reproduce
    • Potential impact
    • Suggested fix (if any)

What to Expect

  • Acknowledgment: Within 48 hours
  • Initial Assessment: Within 7 days
  • Resolution Timeline: Depends on severity
    • Critical: 24-72 hours
    • High: 7 days
    • Medium: 30 days
    • Low: Next release

Scope

This security policy applies to:

  • The AgentScope CLI tool
  • Official npm packages
  • This GitHub repository

Out of Scope

  • Third-party dependencies (report to their maintainers)
  • User misconfiguration
  • Social engineering attacks

Security Design Principles

AgentScope follows these security principles:

  1. No Code Execution - Only reads and parses configuration files
  2. No Network Access - All operations are local
  3. No Secrets Handling - Does not parse or expose API keys
  4. Path Validation - Prevents directory traversal attacks
  5. Input Sanitization - All parsed content is sanitized before output

Security Checklist for Contributors

When contributing, ensure:

  • No hardcoded secrets or credentials
  • No execution of user-provided code
  • Input validation on all external data
  • Path traversal prevention on file operations
  • Dependencies are from trusted sources
  • No verbose error messages exposing internals

There aren’t any published security advisories