response-compression
Pass
Audited by Gen Agent Trust Hub on Feb 17, 2026
Risk Level: SAFENO_CODE
Full Analysis
- NO_CODE (SAFE): The skill contains no executable scripts, tool definitions, or binary files. It consists entirely of Markdown instructions and guidelines.
- PROMPT_INJECTION (SAFE): While the skill contains behavioral instructions (e.g., 'Eliminate response bloat', 'Avoid Trailing Content'), these are standard formatting and style guidelines. There are no attempts to bypass safety filters, extract system prompts, or override core safety protocols.
- DATA_EXFILTRATION (SAFE): There are no network calls, file system operations, or hardcoded credentials detected.
- EXTERNAL_DOWNLOADS (SAFE): No external dependencies or remote script executions are present.
Audit Metadata