Unverified50% confidenceFactExact time
Having the model run commands that produce large amounts of output (such as git status) pollutes the context window and consumes additional tokens.
1
Sources
50%
Confidence
Long-term
Relevance
9/11/2026
First Seen
Sources
Related Entities
Related Claims
Unverified大模型API版本迭代可能伴随输出格式变化、系统提示词响应行为漂移及函数调用/结构化输出接口规范调整，即'行为漂移'现象67% similarUnverified许多模型对System Prompt有特殊处理机制，其权重通常高于普通对话内容，将核心规则写入系统提示比放在普通用户消息中更能抵抗衰减65% similarUnverifiedIn large model usage, the context window is a limited resource, and token consumption accumulates with conversation turns, file reads, and code injection.65% similarUnverified在Cursor使用场景中，发送给模型的上下文包括编辑器自动附加的代码上下文、文件结构信息和项目索引数据，隐含Token消耗往往是用户可见输入的数倍甚至数十倍64% similarUnverifiedIn programming assistant scenarios, system prompts and code context typically consume a large number of tokens, resulting in an extremely high cache hit rate.64% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/902407API
curl https://kongchang.com/api/v1/knowledge/claims/902407MCP
get_claim(id=902407)