hooked-model
Pass
Audited by Gen Agent Trust Hub on Feb 17, 2026
Risk Level: SAFENO_CODE
Full Analysis
- [SAFE] (SAFE): A comprehensive security audit of all 19 files was performed, and no malicious patterns, vulnerabilities, or exploits were detected.
- [NO_CODE] (SAFE): The skill is entirely composed of Markdown documentation and YAML frontmatter. It lacks any executable files (.py, .js, .sh), binary data, or dependency manifests (requirements.txt, package.json), eliminating the possibility of code-based attacks.
- [PROMPT_INJECTION] (SAFE): The instructions are limited to educational frameworks for product design. No directives were found that attempt to override system behavior, bypass safety guardrails, or extract the agent's internal prompt.
- [DATA_EXFILTRATION] (SAFE): There are no patterns involving access to sensitive local file paths (e.g., SSH keys, credentials) or network-reaching commands (curl, wget, fetch) that could lead to data exposure.
Audit Metadata