marketing-kuaishou-strategist
Pass
Audited by Gen Agent Trust Hub on Apr 7, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: No malicious patterns detected in the skill instructions or associated scripts. The skill follows best practices for persona definition and task guidance without attempting to bypass safety filters or exfiltrate data.
- [COMMAND_EXECUTION]: The
scripts/run.shfile provides a standard execution wrapper for the agent skill. It calls a local Python runner (run-skill.py) located in the repository root. No arbitrary or dangerous shell commands were detected. - [DATA_EXFILTRATION]: No network calls to untrusted domains or attempts to read sensitive system files (like SSH keys or credentials) were found. The skill operates within its intended marketing context.
- [PROMPT_INJECTION]: The instructions do not contain attempts to override the system prompt, ignore safety guidelines, or extract developer secrets.
- [SAFE]: Regarding potential indirect prompt injection (Category 8), the skill intends to process market research and competitor data. While this is an ingestion surface, the skill does not exhibit exploitable capabilities like arbitrary command execution on user input, and the intended use case is standard analysis.
Audit Metadata