Skip to content

Security: strands-compose/bedrock-agentcore

Security

SECURITY.md

Security Policy

Supported Versions

Version Supported
0.1.x

Reporting Security Issues

We take security seriously. Please do not open a public GitHub issue for security vulnerabilities.

Instead, please report via GitHub Security Advisories.

Include:

  • Description of the vulnerability
  • Steps to reproduce
  • Potential impact
  • Suggested fix (if any)

Security Practices

strands-compose-agentcore enforces the following (via Bandit + code review):

  • No eval() or exec() — config is parsed through strands-compose's Pydantic pipeline, never executed as code
  • No subprocess with shell=True — all external commands use direct invocation
  • No hardcoded secrets — all credentials resolved from environment variables or AWS IAM
  • No pickle — serialization uses JSON only
  • Strict input validation — all YAML config validated through strands-compose's Pydantic schema
  • Bandit scanning — automated static security analysis (uv run just check-security)
  • Dependency auditing — dependencies are pinned and regularly reviewed for known vulnerabilities

There aren’t any published security advisories