docs: cite public agent loop effectful program lecture (#2927)

Author: huangruiteng

docs/architecture/rfcs/agent-loop-effect-interpreter-v0.md

@@ -24,6 +24,7 @@ interprets each effect request and returns an observation to the next model
 step.
 
 The framing builds on the public lecture series by 齐梦星空:
+[主线一：Agent Loop 是 effectful program(1)](https://www.xiaohongshu.com/discovery/item/6a01d501000000003700c5de?source=webshare&xhsshare=pc_web&xsec_token=ABqpNuladcxhev099wLKw8M3ilhKBua0BQXNpxnBZEGkc=&xsec_source=pc_share),
 [主线一：Tool Calling 是 Kleisli arrow(2)](https://www.xiaohongshu.com/discovery/item/6a02f388000000003502b2d6?source=webshare&xhsshare=pc_web&xsec_token=ABHcIpzpd2RlhAaRr9sZZ-q1OIfRgt7rvG2jn7GUO3tNo=&xsec_source=pc_share)
 and
 [主线一：Agent Loop 里的小魔法：函数的组合(3)](https://www.xiaohongshu.com/discovery/item/6a057524000000003701f6aa?source=webshare&xhsshare=pc_web&xsec_token=AB43lNCJ5ULmfTrGfeTLWd2-jQ6q8nFMGyNAd-tlXJ1uw=&xsec_source=pc_share).
@@ -447,6 +448,8 @@ story explicit and gives the refactor and test work a stable target.
 
 ## References
 
+- 齐梦星空,
+  [*主线一：Agent Loop 是 effectful program(1)*](https://www.xiaohongshu.com/discovery/item/6a01d501000000003700c5de?source=webshare&xhsshare=pc_web&xsec_token=ABqpNuladcxhev099wLKw8M3ilhKBua0BQXNpxnBZEGkc=&xsec_source=pc_share).
 - 齐梦星空,
   [*主线一：Tool Calling 是 Kleisli arrow(2)*](https://www.xiaohongshu.com/discovery/item/6a02f388000000003502b2d6?source=webshare&xhsshare=pc_web&xsec_token=ABHcIpzpd2RlhAaRr9sZZ-q1OIfRgt7rvG2jn7GUO3tNo=&xsec_source=pc_share).
 - 齐梦星空,

docs/development/control-plane-course/01-agent-loop-effectful-program.md

@@ -11,6 +11,7 @@ interpreter。agent loop 是被解释的循环，harness 才是解释外部 effe
 全文参考：
 [Agent Loop Effect Interpreter RFC](../../architecture/rfcs/agent-loop-effect-interpreter-v0.md)。
 公开来源：
+[主线一：Agent Loop 是 effectful program(1)](https://www.xiaohongshu.com/discovery/item/6a01d501000000003700c5de?source=webshare&xhsshare=pc_web&xsec_token=ABqpNuladcxhev099wLKw8M3ilhKBua0BQXNpxnBZEGkc=&xsec_source=pc_share)、
 [主线一：Tool Calling 是 Kleisli arrow(2)](https://www.xiaohongshu.com/discovery/item/6a02f388000000003502b2d6?source=webshare&xhsshare=pc_web&xsec_token=ABHcIpzpd2RlhAaRr9sZZ-q1OIfRgt7rvG2jn7GUO3tNo=&xsec_source=pc_share)、
 [主线一：Agent Loop 里的小魔法：函数的组合(3)](https://www.xiaohongshu.com/discovery/item/6a057524000000003701f6aa?source=webshare&xhsshare=pc_web&xsec_token=AB43lNCJ5ULmfTrGfeTLWd2-jQ6q8nFMGyNAd-tlXJ1uw=&xsec_source=pc_share)。
 

docs/product/core-control-plane/state-machine.md

@@ -68,7 +68,8 @@ This is the model described in the
 [Agent Loop Effect Interpreter RFC](../../architecture/rfcs/agent-loop-effect-interpreter-v0.md).
 The agent loop is the loop. The harness is the effectful program. The state
 machine is not the product; it is the decision table inside that effect
-interpreter.
+interpreter. This framing follows the public lecture
+[主线一：Agent Loop 是 effectful program(1)](https://www.xiaohongshu.com/discovery/item/6a01d501000000003700c5de?source=webshare&xhsshare=pc_web&xsec_token=ABqpNuladcxhev099wLKw8M3ilhKBua0BQXNpxnBZEGkc=&xsec_source=pc_share).
 
 | State family | Input effect | Interpreter | Decision | Observation | Next effect |
 |---|---|---|---|---|---|