Unverified80% confidenceOpinionTime unknown
AI编程助手如Claude Code在没有环境检测工具辅助时，会因训练分布偏差默认选择统计上最常见的工具（如npm而非pnpm）
1
Sources
80%
Confidence
Long-term
Relevance
5/31/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedCurrent training data audit practices primarily rely on developer self-disclosure (such as Meta's Llama series and Stability AI's model cards), lacking independent and technically verifiable means68% similarUnverifiedAlignment training optimizes the model's statistical behavior in known scenarios rather than establishing formally verifiable security boundaries, and may fail when faced with rare instruction combinations or role-playing frameworks outside the training distribution.68% similarUnverifiedLLM training data covers a large amount of open-source code, but this represents statistical patterns rather than an understanding of specific projects.67% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/12134API
curl https://kongchang.com/api/v1/knowledge/claims/12134MCP
get_claim(id=12134)