Expired50% confidenceBenchmarkExact time
Opus performed the worst in terms of tool misuse, while Opus 5 and 4.8 were the most aggressive in over-building.
1
Sources
50%
Confidence
Short-term (~14 days)
Relevance
8/25/2026
First Seen
Valid until: 9/8/2026(expired)
Sources
Related Entities
Related Claims
UnverifiedIn Theo's history audit, Opus 5 aggressively killed the wrong processes, killing more processes in two days than he had with all other tools combined.61% similarUnverifiedOpus 5擅长发现漏洞但利用漏洞的能力被大幅削弱，属于内化式对齐60% similarUnverified工具过于细粒度会导致规划步骤爆炸，工具过于粗粒度则会牺牲灵活性，这是Agent系统设计中最常见的权衡问题之一58% similarUnverifiedClaude Opus 4.7 performs poorly on complex backend tasks, falling short of expectations in scenarios such as API refactoring, architecture design, database logic, and cross-file debugging.58% similarUnverifiedThe hallucinated-field problem is more severe on newer Anthropic models (Opus 4.8 and Sonnet 5) than on older models.58% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/797940API
curl https://kongchang.com/api/v1/knowledge/claims/797940MCP
get_claim(id=797940)