agentic-workflow-advisor
Pass
Audited by Gen Agent Trust Hub on Oct 15, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill utilizes a local Python script (
extract_flow_info.py) to parse workflow structures. This script performs static analysis using the standardjsonandastlibraries, which is a secure and necessary function for the skill's stated purpose. - [INDIRECT_PROMPT_INJECTION]: The skill is designed to process untrusted workflow definitions provided by users. It mitigates potential injection risks by implementing strict ingestion rules—instructing the agent to wait for explicit user input and avoid unauthorized disk access—and by relying on static parsing rather than dynamic execution of the analyzed content.
Audit Metadata