Tool-using LLM agents

Named by 2 kept items in this workspace.

Search every workspace for Tool-using LLM agents. It matches the name as written, so it can find other things that share it.

The items that name it 2

RAISED: Self-Distillation for Robustness to Prompt Injection in LLM Agents
paper2026-10-05
Agents that act on untrusted external content (tool outputs) are vulnerable to indirect prompt injection
Compromise Is Not Consequence: Evaluating Task-Scoped Authorization in LLM Agents with Paired Replay
paper2026-10-05It is written here as Tool-using LLM agents (five local model configurations, across four tool domains).
five local model configurations, across four tool domains

All technologies in this workspace.