Unverified50% confidenceFactExact time
On Sonnet 5.5, of the 20 skills, 12 were clearly helpful, 4 were clearly worse, and 4 had insufficient samples to judge.
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
10/19/2026
First Seen
Valid until: 1/17/2027
Sources
Related Entities
Related Claims
UnverifiedOn Opus 5.5, of the 20 skills, 12 were clearly helpful, 2 were clearly worse, and 6 had insufficient samples to judge.80% similarUnverifiedThe criteria are divided into four families—foundational, premature disclosure, goal deviation, and information omission—with 19 failure modes in total, of which 18 were triggered in actual testing.71% similarUnverifiedOf the 21 outputs, 14 were judged as "too technical," and 13 were flagged as lacking some capability.70% similarUnverifiedWhen the benchmark set grew from 10 tasks to 20, an agent that originally passed all 10/10 showed 10 failures under the new counting method, causing it to be misjudged as capability decay and automatically quarantined.69% similarUnverified在工具数量超过20个的场景下，工具描述的语义歧义度与模型选择准确率之间存在显著的负相关关系65% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1065261API
curl https://kongchang.com/api/v1/knowledge/claims/1065261MCP
get_claim(id=1065261)