The LLM-agent ecosystem is rapidly growing around reusable skills, making skill reuse auditing increasingly important. Existing code clone detection methods struggle with multi-modality skill reuse. We introduce SKILLTRACE, a multi-trace provenance auditing framework for LLM-agent skill reuse. SKILLTRACE extracts three provenance traces: Expression, Implementation, and Operational, representing the Operational Trace as a Skill Operational Graph (SOG) that captures activation, procedure, and resource-flow structure. On the SKILLTRACE-BENCH benchmark, SKILLTRACE achieves an AUROC of 0.938 and an F1 score of 0.898. Blogger's Review: SKILLTRACE provides an effective solution for LLM-agent skill reuse auditing, with its multi-trace provenance auditing framework offering better detection of skill reuse, having a wide range of application prospects and research value.