migrating-to-workflow-sdk
Pass
Audited by Gen Agent Trust Hub on Sep 14, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- Standardized Migration Patterns (SKILL.md, references/shared-patterns.md): The skill establishes clear rules for code transformation, ensuring that side effects, orchestration, and retries are handled through dedicated SDK primitives. This structure promotes predictable and secure behavior in the migrated workflows.
- Secure Configuration Practices (references/shared-patterns.md, references/aws-step-functions.md): Example code snippets demonstrate the use of environment variables (e.g., process.env.ORDER_API_URL) for external service URLs and identifiers, which is the standard approach for preventing hardcoded secrets.
- Idempotency and Reliability (SKILL.md, references/retries.md): The skill mandates the use of getStepMetadata().stepId for external writes, helping to prevent unintended side effects during workflow retries—a key reliability and security feature in distributed systems.
- Indirect Prompt Injection Surface (SKILL.md): As a tool that transforms user-provided code, the skill is subject to potential indirect prompt injection if the source code contains malicious instructions. However, the instructions provide the agent with a detailed verification checklist to ensure the resulting code adheres strictly to the target SDK's requirements, reducing the risk of the agent being misled.
Audit Metadata