HarmURLBench

What is HarmURLBench?

AgentREVEAL is a diagnostic framework that evaluates safety alignment degradation in LLM agents when web retrieval is integrated. It assesses the impact of…

Released

2026-05-28

Evaluates

Cybersecurity, Information retrieval, Safety, Safety & Trustworthiness

Openness

unknown

Importer

Claire Radar

Review status

unreviewed

Reported scores

0

Source provenance

• Original evidence https://arxiv.org/abs/2605.29224

HarmURLBench paper

• Paper https://arxiv.org/abs/2605.29224

Which sources cite HarmURLBench?

• HarmURLBench

Related Cybersecurity, Information retrieval, Safety, Safety & Trustworthiness benchmarks

• ActBench
• AdversaBench
• Agent Planning Benchmark
• Agent Retrieval Bench
• AgentHarm
• AgentHazard
• AgentS4D
• AI Security Leaderboard