← back to paper

Review history

arxiv: 2607.11098 · 2 revisions

AgentCheck: A Reproduce-Intervene-Mitigate Workbench for LLM Agents over MCP

1. 2026-07-15 UNVERDICTED LOW v1.1.0-grok45 novelty 6.0
   34310 ms 6168 in 2681 out 2026-07-15T09:00:03.797122+00:00
2. 2026-07-14 ACCEPT HIGH v1.1.0-grok45 novelty 6.0
   50529 ms 18366 in 3759 out 2026-07-14T07:04:44.430888+00:00