Unverified50% confidenceSolutionExact time
In the Agent loop, the model is called each round; if the model returns a tool call request, the system executes it and returns the result; if the model gives an answer directly, the loop terminates, and the loop has a maximum number of rounds.
1
Sources
50%
Confidence
Long-term
Relevance
10/29/2026
First Seen
Sources
Agentic RAG in Practice: From Principles to Code, Building an Intelligent Retrieval System Step by Step
bilibiliAgent智能体_10/16/2026
Related Entities
Related Claims
Unverified在Agent Loop中如果只将工具结果塞回messages而漏掉assistant消息,模型会陷入无限循环,不断重复调用工具79% similarUnverifiedA single LLM API call is an instantaneous question-and-answer cycle that ends once the request is sent and the response is returned, whereas an Agent Loop is a continuously running loop structure that persists until the goal is achieved or an exit condition is triggered.77% similarUnverified驱动Agent Loop循环的两个信号来自模型返回的stop_reason字段:当stop_reason等于tool_use时宿主执行工具并继续循环,否则循环退出76% similarUnverifiedThe termination condition of an agentic loop typically falls into one of two categories: the model returns a plain-text response (stop_reason is end_turn), or the preset maximum number of iterations is reached.76% similarUnverifiedA run represents the complete lifecycle from the user's request to the final answer, and may contain multiple turns; each turn is one cycle of "model reasoning—tool call—result backfill."75% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1110359API
curl https://kongchang.com/api/v1/knowledge/claims/1110359MCP
get_claim(id=1110359)