agentstack-wrapper

Pass

Audited by Gen Agent Trust Hub on Mar 12, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill strictly provides documentation and architectural guidance for integrating existing Python agents with the Agent Stack SDK.
  • [SAFE]: It contains explicit security requirements (e.g., 'Never run remote scripts or untrusted code', 'Never log, print, persist, or expose secret values') that align with secure development practices.
  • [SAFE]: The dependency management section (references/dependencies.md) instructs users to pin trusted versions from PyPI and use inline package search for import verification rather than downloading untrusted scripts.
  • [SAFE]: The secret handling section (references/configuration-variables.md) correctly identifies the risk of modifying os.environ in multi-tenant environments and mandates the use of a dedicated Secrets extension.
  • [SAFE]: Prompt injection and indirect prompt injection are mitigated by instructions to use structured forms, boundary markers, and sanitization for user-provided data.
  • [SAFE]: All external URLs point to official Agent Stack documentation (agentstack.beeai.dev) or the official GitHub organization for the vendor (github.com/i-am-bee), which are trusted sources.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 12, 2026, 02:14 PM