chain
Pass
Audited by Gen Agent Trust Hub on Sep 14, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill provides a framework for multi-step tool workflows (chains) where data flows between discrete tools, creating an inherent attack surface for indirect prompt injection if untrusted data is processed.
- Ingestion points: Defined in the "Data Flow" section of
SKILL.md, where output from one step serves as input for the next. - Boundary markers: The design template does not explicitly mandate delimiters or "ignore embedded instructions" markers between connected tool steps.
- Capability inventory: This skill is a structural guide and does not possess intrinsic capabilities for file access, network operations, or shell execution.
- Sanitization: The template does not define specific sanitization or validation logic for the data passing between tool links.
- [SAFE]: No other threat patterns, such as obfuscation, credential exfiltration, or unauthorized command execution, were detected. The skill follows established patterns for modular agent workflow design.
Audit Metadata