update frozenlake readme

cookbook/frozenlake/frozenlake_react_agent.py

@@ -68,10 +68,7 @@ class FrozenLakeReactAgent:
                 return yaml.safe_load(f)
         except FileNotFoundError:
             logger.warning("Prompt file not found, using default prompts")
-            return {
-                "frozenlake_sys_prompt_no_slippery": "You are playing FrozenLake. Navigate from S to G avoiding H.",
-                "frozenlake_sys_prompt_slippery": "You are playing FrozenLake. Navigate from S to G avoiding H. Ice is slippery!"
-            }
+            raise FileNotFoundError("Prompt file not found. Please check your current path (should be ./cook/frozenlake) and try again.")
 
     def call_llm(self, messages: List[Dict]) -> str:
         """Call LLM with retry logic"""
@@ -158,8 +155,7 @@ class FrozenLakeReactAgent:
             r'["\']action["\']\s*:\s*["\']([0-3])["\']',
             r'"action"\s*:\s*"([0-3])"',
             r"'action'\s*:\s*'([0-3])'",
-            r'\baction["\']?\s*[:=]\s*["\']?([0-3])',
-            r'\b([0-3])\b(?=\s*$)',  # Single digit at end
+            r'\baction["\']?\s*[:=]\s*["\']?([0-3])'
         ]
 
         for pattern in patterns:
@@ -223,6 +219,7 @@ class FrozenLakeReactAgent:
         for step in range(self.max_steps):
             # Get action from LLM
             response = self.call_llm(messages)
+            logger.info(response)
             action = self.action_parser(response)
 
             messages.append({"role": "assistant", "content": response})

cookbook/frozenlake/quickstart.md

@@ -51,10 +51,13 @@ reme \
   vector_store.default.backend=local
 ```
 
-Load default memory library for FrozenLake:
+Add your api key for agent:
 ```bash
+export OPENAI_API_KEY="xxx"
+export OPENAI_BASE_URL="xxx"
 ```
 
+
 ## Run Experiments
 
 ### 1. Quick Test: Performance Evaluation Only (Default)
@@ -62,6 +65,7 @@ Load default memory library for FrozenLake:
 Run the main experiment script to test agent performance using existing memory:
 
 ```bash
+cd cookbook/frozenlake
 python run_frozenlake.py
 ```
 
@@ -131,14 +135,6 @@ python run_exp_statistic.py
    - Calculates performance metrics
    - Generates comparative analysis
 
-## Understanding Results
-
-The experiment evaluates agent performance on FrozenLake maps:
-
-- **Success Rate**: Percentage of episodes where the agent reaches the goal
-- **With vs. Without Memory**: Compares performance with and without task memory
-- **Slippery vs. Non-slippery**: Compares performance in different environment dynamics
-
 ### Output Files
 
 - `./exp_result/*_training.jsonl`: Results from training phase