Recursive language models for jailbreak detection: a procedural defense for tool-augmented agents

source_context

{
  "surface": "signal_canvas",
  "mode": "paper",
  "query": "Recursive language models for jailbreak detection: a procedural defense for tool-augmented agents",
  "normalized_query": "2602.16520",
  "route": "/signal-canvas/recursive-language-models-for-jailbreak-detection-a-procedural-defense-for-tool-augmented-agents",
  "paper_ref": "recursive-language-models-for-jailbreak-detection-a-procedural-defense-for-tool-augmented-agents",
  "topic_slug": null,
  "benchmark_ref": null,
  "dataset_ref": null
}