Agent Workflow Builder
Pass
Audited by Gen Agent Trust Hub on Mar 30, 2026
Risk Level: SAFEPROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
- [PROMPT_INJECTION]: The skill provides code templates for agent loops that define an attack surface for indirect prompt injection.\n
- Ingestion points: The
user_inputvariable in theAgent.runmethod inSKILL.mdis interpolated directly into the message list.\n - Boundary markers: No delimiters or safety instructions are included in the template to separate user input from system instructions.\n
- Capability inventory: The
execute_toolmethod allows the agent to trigger external capabilities based on LLM reasoning.\n - Sanitization: No input validation or escaping mechanisms are implemented in the provided Python examples.\n- [EXTERNAL_DOWNLOADS]: The skill's metadata includes an installation command that fetches the markdown definition from a remote repository on GitHub.
Audit Metadata