template-engineering
Pass
Audited by Gen Agent Trust Hub on Feb 16, 2026
Risk Level: LOWNO_CODE
Full Analysis
- SAFE (SAFE): The skill is entirely instructional, providing a framework for engineers to design prompt templates. No malicious instructions or security bypasses were detected.
- NO_CODE (SAFE): The skill contains no executable scripts or code logic. It defines allowed tools for searching and reading files (Read, Grep, Glob) which are safe within the context of a documentation and planning aid.
- INDIRECT_PROMPT_INJECTION (INFO): While the guide illustrates how to use parameters like '$ARGUMENTS' which can ingest untrusted data, the skill itself does not implement these templates or execute actions on untrusted inputs. It provides guidance on creating structured 'Plan' formats which is a recommended mitigation against injection.
Audit Metadata