Skip to content

Security Findings from Trustabl Scanner #168

Description

@joshua-trustabl

We ran the Trustabl scanner on your repo and discovered several medium severity findings. In contributing/samples/agent_governance/main.py, the LlmAgent lacks a description, which could affect routing delegation between agents in Google ADK. Additionally, both contributing/samples/agent_governance/main.py and contributing/samples/hitl_approval/credit_agent/agent.py do not specify safety_settings for their agents, meaning that Gemini models' content filters default to OFF. These findings could potentially impact the security and functionality of your agents in Google ADK. Please review these areas to ensure they meet your requirements.


Add Trustabl to your CI — trustabl/trustabl-action:

- name: Trustabl scan
  uses: trustabl/trustabl-action@v1

Metadata

Metadata

Assignees

No one assigned

    Labels

    No labels
    No labels

    Type

    No type

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions