Fixed to be able to parse tasks even when rules are recited; added instructions on how to stop containers in README.md

Author: saten-private

README-jp.md

@@ -75,13 +75,19 @@ https://x.com/saten_work/status/1791558481432232355
 
 前提条件として、docker と docker-compose がインストールされている必要があります。Docker desktop は最もシンプルなオプションです https://www.docker.com/products/docker-desktop/
 
-以下を実行してください。
+## 実行
 
 ```
 docker-compose up -d && docker attach babyagi
 ```
 
-**注意：Ctrl+Cで終了しても```./clean.sh```の実行やDocker Desktopなどでコンテナ停止しないと停止しなくなりました。ご注意ください。**
+## 停止
+
+```
+docker-compose stop
+```
+
+**注意：Ctrl+Cで終了しても```docker-compose stop```や```./clean.sh```を実行しない限り停止しません。ご注意ください。**
 
 **注意：目的を達成できず、ループし続けることがあります。AnthropicやOpenAIのAPIの使用料にご注意ください。**
 

README.md

@@ -75,13 +75,19 @@ Please follow the steps below:
 
 As a prerequisite, docker and docker-compose must be installed. Docker desktop is the simplest option https://www.docker.com/products/docker-desktop/
 
-Please execute the following:
+## Run
 
 ```
 docker-compose up -d && docker attach babyagi
 ```
 
-**Note: Even if you exit with Ctrl+C, it will not stop unless you run ```./clean.sh``` or stop the container using Docker Desktop. Please be careful.**
+## Stop
+
+```
+docker-compose stop
+```
+
+**Note: Even if you exit with Ctrl+C, it will not stop unless you run ```docker-compose stop``` or ```./clean.sh```. Please be careful.**
 
 **Note: The agent might loop indefinitely if it cannot achieve its objective. Please be aware of the cost of Anthropic and OpenAI APIs usage.**
 

task_parser.py

@@ -4,7 +4,15 @@
 class TaskParser:
 
     def decode(self, input_string: str) -> List[dict]:
-        data = input_string.strip().split("type:")[1:]
+        input_string = input_string.strip()
+        if input_string.startswith("type:"):
+          # The "type:" at the beginning should be deleted because it will all be added later.
+          input_string = input_string[len("type:"):]
+          data = input_string.split("\ntype:")
+        else:
+          # Ignore the first part of the "\ntype:" division if it does not begin with "type:".
+          data = input_string.split("\ntype:")[1:]
+
         if len(data) == 0:
             raise ValueError("No valid items found")
         parsed_data = []