go-style-guide
Fail
Audited by Gen Agent Trust Hub on Feb 16, 2026
Risk Level: HIGHPROMPT_INJECTIONNO_CODE
Full Analysis
- PROMPT_INJECTION (HIGH): The skill is susceptible to Indirect Prompt Injection due to its interaction with untrusted external code. * Ingestion points: The skill processes user-provided Go source code, concurrency patterns, and performance data. * Boundary markers: The skill lacks instructions to help the agent distinguish between Go code logic and malicious natural language instructions embedded within that code. * Capability inventory: The skill is intended for refactoring and writing Go code, which involves high-privilege file system modifications. * Sanitization: There is no requirement or logic for the agent to sanitize or escape external code content before performing actions.
- NO_CODE (LOW): This skill contains only markdown documentation and guidelines; it does not include any executable scripts or source code files.
Recommendations
- AI detected serious security threats
Audit Metadata