agent-expertise-protocol
Pass
Audited by Gen Agent Trust Hub on Nov 5, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill provides purely instructional content for agent governance and does not contain any executable scripts, commands, or network operations. All referenced resources are relative paths within the vendor's own repository structure.
- [PROMPT_INJECTION]: No patterns of prompt injection, role-play overrides, or safety filter bypasses were detected. The instructions reinforce safety and user control through an explicit authority model.
- [DATA_EXPOSURE]: There are no hardcoded credentials, API keys, or access to sensitive local file paths like SSH keys or environment files.
- [REMOTE_CODE_EXECUTION]: The skill does not perform any external downloads, package installations, or execution of code from remote sources.
- [OBFUSCATION]: No techniques such as Base64 encoding, zero-width characters, or homoglyphs are used to hide content. The document is clear and readable.
Audit Metadata