writing-skills
Pass
Audited by Gen Agent Trust Hub on Jul 26, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill is composed exclusively of documentation files (SKILL.md, influence-principles.md, and pressure-testing.md) that provide guidelines for writing predictable agent behaviors and established testing protocols. No malicious patterns, obfuscation, or security vulnerabilities were detected.
- [PROMPT_INJECTION]: The content instructs developers on how to build skills that resist agent rationalization and adhere to defined rules. No techniques for bypassing safety filters, overriding system instructions, or jailbreaking were found.
- [DATA_EXFILTRATION]: The skill does not involve any file system access to sensitive directories, hardcoded credentials, or network requests to external domains.
- [REMOTE_CODE_EXECUTION]: There is no evidence of external package installation, remote script execution, or dynamic code evaluation in the provided skill set.
Audit Metadata