🇺🇸

Author: KillianLucas

interpreter/core/core.py

@@ -77,7 +77,7 @@ def __init__(
         import_computer_api=False,
         skills_path=None,
         import_skills=False,
-        multi_line=False,
+        multi_line=True,
         contribute_conversation=False,
     ):
         # State

interpreter/terminal_interface/start_terminal_interface.py

@@ -278,7 +278,17 @@ def start_terminal_interface(interpreter):
             sys.argv.remove(old_flag)
             sys.argv.append(new_flag)
 
-    parser = argparse.ArgumentParser(
+    class CustomHelpParser(argparse.ArgumentParser):
+        def print_help(self, *args, **kwargs):
+            super().print_help(*args, **kwargs)
+            special_help_message = '''
+Open Interpreter, 2024
+
+Use """ to write multi-line messages.
+            '''
+            print(special_help_message)
+
+    parser = CustomHelpParser(
         description="Open Interpreter", usage="%(prog)s [options]"
     )
 

interpreter/terminal_interface/utils/cli_input.py

@@ -1,6 +1,6 @@
 def cli_input(prompt: str = "") -> str:
-    start_marker = "```"
-    end_marker = "```"
+    start_marker = '"""'
+    end_marker = '"""'
     message = input(prompt)
 
     # Multi-line input mode

scripts/wtf.py

@@ -7,7 +7,9 @@
 import platform
 import time
 
+import platformdirs
 import pyperclip
+import yaml
 from pynput.keyboard import Controller, Key
 
 os.environ["LITELLM_LOCAL_MODEL_COST_MAP"] = "True"
@@ -21,7 +23,7 @@
 1. Quickly scan the provided terminal history.
 2. Identify the most recent error or issue.
 3. Determine the most likely solution or debugging step.
-4. Respond with a brief explanation followed by a markdown code block containing a shell command to address the issue.
+4. Respond with a VERY brief explanation followed by a markdown code block containing a shell command to address the issue.
 
 Rules:
 - Provide a single shell command in your code block, using line continuation characters (\\ for Unix-like systems, ^ for Windows) for multiline commands.
@@ -31,6 +33,7 @@
 - Place explanatory text before the code block.
 - NEVER USE COMMENTS IN YOUR CODE.
 - Focus on the most recent error, ignoring earlier unrelated commands.
+- The error may be as simple as a spelling error, or as complex as requiring tests to be run. You may even need to suggest a command that edits text in a file to fix some code.
 - Prioritize speed and conciseness in your response. Don't use markdown headings. Don't say more than a sentence or two. Be incredibly concise.
 
 User's System: {platform.system()}
@@ -107,6 +110,18 @@ def main():
         {"role": "user", "content": history},
     ]
 
+    # Get LLM model from profile
+    default_profile_path = os.path.join(
+        platformdirs.user_config_dir("open-interpreter"), "profiles", "default.yaml"
+    )
+
+    try:
+        with open(default_profile_path, "r") as file:
+            profile = yaml.safe_load(file)
+            model = profile.get("llm", {}).get("model", "gpt-3.5-turbo")
+    except:
+        model = "gpt-3.5-turbo"
+
     # Process LLM response
     in_code = False
     backtick_count = 0
@@ -115,7 +130,7 @@ def main():
     started = False
 
     for chunk in litellm.completion(
-        model="gpt-3.5-turbo", messages=messages, temperature=0, stream=True
+        model=model, messages=messages, temperature=0, stream=True
     ):
         if not started:
             started = True