AgentTrace: Execution-Level Fingerprinting of Black-Box LLM Agents.

arXiv:2610.32092v1 Announce Type: cross Abstract: Verifying the provenance of autonomous LLM agents is a critical security challenge. Existing semantic attribution is easily obfuscated by system prompts, while active watermarking conflicts with schema-constrai