Unverified50% confidenceSolutionExact time
In the scoring bug, the original strict comparison often equated an application's inability to parse directly with an incorrect model answer — for example, JSON wrapped in a Markdown code fence would be rejected by the JSON parser even if the values were correct.
1
Sources
50%
Confidence
Long-term
Relevance
10/12/2026
First Seen
Sources
Related Entities
Related Claims
Unverified隐形字符可能导致字符串比较失败、正则匹配失效、JSON解析错误或数据库查询返回空结果等难以调试的bug73% similarUnverified工具调用的可靠性高度依赖模型输出严格遵循预定义的 JSON Schema，任何字段类型偏差、缺失或多余都可能导致框架层的解析失败72% similarUnverifiedIn this case, the JSON parsing exception is most likely merely a symptom of the failure rather than the root cause itself69% similarUnverifiedOnce final JSON parsing fails, subsequent semantic checks are skipped entirely, which is the technical root cause of the 'task completed yet zero score' phenomenon.69% similarUnverifiedThe most common model failure is returning something that is 'almost JSON,' such as extra commas, markdown code fences, verbose preambles, or writing the number 3 as the word 'three,' with the AI node reporting success while the workflow crashes only several steps later when parsing fails.69% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1003726API
curl https://kongchang.com/api/v1/knowledge/claims/1003726MCP
get_claim(id=1003726)