Unverified50% confidenceFactExact time
Prompt engineering and RAG are passive question-answering modes: if the user does not ask, nothing is output, and tasks cannot be executed proactively.
1
Sources
50%
Confidence
Long-term
Relevance
10/23/2026
First Seen
Sources
Mastering AI Agent Development: From Underlying Logic to Enterprise-Level Practice
bilibili图灵程序员10/23/2026
Related Entities
Related Claims
UnverifiedIn RAG and Agent systems, prompts not only carry the user's question but also need to embed retrieved context, tool-calling formats and historical dialogue, and the quality of prompt engineering directly affects system performance.68% similarUnverifiedAsk mode is for pure question-and-answer interactions and will not proactively modify any project files.65% similarUnverifiedMost traditional LLM applications adopt a request-response model driven by immediate user input, where the question of when to execute does not arise.64% similarUnverifiedThe Know layer injects policies, constraints, and business rules into the agent through system prompts or RAG, but does not enforce restrictions.64% similarUnverifiedAsk User Question 本质上是一个特殊的 Tool，当模型调用它时框架会暂停 Agent 执行循环，将问题展示给用户等待输入后再注入上下文继续执行61% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1091778API
curl https://kongchang.com/api/v1/knowledge/claims/1091778MCP
get_claim(id=1091778)