IFAS study: how tool/API failures and task-state loss affect the safety behaviour of LLM agents. Active research design.
-
Updated
Oct 5, 2026 - Python
#
toolsandbox
Here are 2 public repositories matching this topic...
Benchmarking schema-valid false tool observations and defense baselines for tool-using LLM agents.
benchmark mcp ai-safety tool-use prompt-injection llm-agents agent-security rag-security agentdojo toolsandbox tool-output-spoofing
-
Updated
Jun 10, 2026 - Python
Add this topic to your repo
To associate your repository with the toolsandbox topic, visit your repo's landing page and select "manage topics."