ag2-use-builtin-tools

Pass

Audited by Gen Agent Trust Hub on May 8, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill provides documentation and recipes for standard functionality of the AG2 framework. No malicious instructions or hidden payloads were found.\n- [COMMAND_EXECUTION]: Documents the use of FilesystemToolkit and CodeExecutionTool. The filesystem tool implementation highlights the use of a base_path parameter to enforce sandboxing and prevent path-traversal attacks.\n- [EXTERNAL_DOWNLOADS]: References the installation of framework extensions (e.g., ag2[exa]) through standard package managers. These are verified vendor resources from the ag2ai ecosystem.\n- [REMOTE_CODE_EXECUTION]: Outlines the usage of MCPServerTool and SkillsToolkit for extending agent capabilities. These are standard integration points within the framework and are presented with appropriate configuration parameters.
Audit Metadata
Risk Level
SAFE
Analyzed
May 8, 2026, 08:42 PM
Security Audit — agent-trust-hub — ag2-use-builtin-tools