Fetching from the wire…
01
02
03
04
05
06
07
08
Agents2026-07-12 · source-backed
A 1,000-task diagnostic benchmark across finance, medicine, law, cybersecurity, and real estate that labels traces as Tool-Skip, Result-Ignore, Output-Fabrication, or Unnecessary-Tool-Use (arXiv). Across 19 frontier models the best hits only 86.33% clean tool-use, and models with similar aggregate scores fail in completely different ways. That last part is the useful bit. A single pass/fail number hides the failure mode you actually need to fix. Use it to profile your agent, not just grade it.
Each link below shares sources, entities, or timing with this story.
output uses Claude Code / Same source domain / Shared topic
Linked by a graph relationship (output uses Claude Code); reported by the same outlet (arxiv.org); overlapping topics (actually, agent, different, failure).
Linked by a graph relationship (output uses Claude Code); reported by the same outlet (arxiv.org); overlapping topics (actually, agent, benchmark, model).
output uses Claude Code / Shared entity: Tool / Shared topic / Earlier coverage
Linked by a graph relationship (output uses Claude Code); both cover Tool; overlapping topics (actually, agent).
Linked by a graph relationship (output uses Claude Code); both cover Tool; overlapping topics (agent, model).
output uses Claude Code / Shared entity: Tool / Same source domain / Earlier coverage
Linked by a graph relationship (output uses Claude Code); both cover Tool; reported by the same outlet (arxiv.org).
output uses Claude Code / Shared topic / Tension
Linked by a graph relationship (output uses Claude Code); overlapping topics (agent, benchmark, best, different); pushes against this story (but).
Shared entity: Tool / Same source domain / Shared topic / Earlier coverage / Tension
Both cover Tool; reported by the same outlet (arxiv.org); overlapping topics (agent, diagnostic).
output uses Claude Code / Same source domain / Shared topic
Linked by a graph relationship (output uses Claude Code); reported by the same outlet (arxiv.org); overlapping topics (agent, benchmark, model).