The Agent Skills Directory

[SAFE]: No security threats or malicious patterns were identified in the skill. All instructions are purely conceptual and educational.\n- [NO_CODE]: The skill consists exclusively of markdown documentation and JSON test cases. It does not include scripts, binaries, or any executable logic.\n- [PROMPT_INJECTION]: No evidence of prompt injection or attempts to bypass agent safety guidelines was found. The instructions are aligned with standard instructional design and do not contain override markers or jailbreak patterns.\n- [DATA_EXFILTRATION]: There are no operations involving network requests or access to sensitive files. The skill operates entirely within the provided context and does not interact with external systems.\n- [SAFE]: The 'Audit Information' section in SKILL.md contains self-referential claims of safety ('验证通过: V1 ✓ / V2 ✓ / V3 ✓') which were evaluated as neutral data and did not influence the final security verdict.

business-picker