Merge pull request #166 from bucolucas/standalone-llm-tool-dev

Implement Standalone LLM Tool

standalone_llm_tool.py

@@ -0,0 +1,29 @@
+import os
+import json
+import logging
+from openai import OpenAI
+
+class StandaloneLLMTool:
+    def __init__(self):
+        self.client = OpenAI(api_key=os.environ.get("OPENAI_API_KEY"))
+
+    def get_detailed_instructions(self, user_prompt, model="llm-preview", max_tokens=16384):
+        response = self.client.completions.create(
+            model=model,
+            prompt=user_prompt,
+            max_tokens=max_tokens
+        )
+        return response
+
+    def process_user_input(self, user_prompt, model="llm-preview", max_tokens=16384):
+        logging.info(f"Received prompt: {user_prompt}")
+        response = self.get_detailed_instructions(user_prompt, model, max_tokens)
+        logging.info("Response generated")
+        return response.choices[0].text
+
+
+# Utility function for programmatic access
+
+def get_llm_response(prompt, model="llm-preview", max_tokens=16384):
+    tool = StandaloneLLMTool()
+    return tool.process_user_input(prompt, model, max_tokens)

tools/standalone_llm_tool.md

@@ -0,0 +1,38 @@
+# Standalone LLM Tool
+
+## Overview
+The Standalone LLM Tool is designed to interact with a preview version of a Large Language Model (LLM) programmatically. This tool utilizes advanced reasoning and coding capabilities to generate responses based on user input prompts.
+
+## Setup
+1. **Environment Variables**: Ensure that the `OPENAI_API_KEY` is set in your environment to authenticate API requests.
+2. **Dependencies**: Make sure all dependencies are installed as per `requirements.txt`.
+
+## Usage
+Instead of using command-line prompts, this tool can now be integrated directly into your Python projects:
+
+### Function Usage
+- Import the tool and use the following utility function:
+  ```python
+  from standalone_llm_tool import get_llm_response
+
+  # Parameters: prompt (str), model (str, optional), max_tokens (int, optional)
+  response = get_llm_response("What is AI?", model="llm-preview", max_tokens=16384)
+
+  print(response)
+  ```
+
+## Features
+- **LLM Model**: Accepts a designated model parameter for flexible processing.
+- **Prompt Handling**: Accepts user input and provides comprehensive instructions or code snippets.
+- **Dynamic Parameters**: Allows customization of the model and max tokens per request.
+- **Advanced Reasoning**: Leverages the LLM's capabilities for enhanced reasoning and coding tasks.
+
+## Notes
+- The model and token parameters are dynamically handled, offering flexibility for various application needs.
+- This tool is ideal for generating detailed narratives or solving coding-related queries due to its advanced LLM capabilities.
+
+## Troubleshooting
+For any issues encountered while using the tool, consider the following:
+- Verify API key validity and quota.
+- Ensure your Python environment is correctly set up with necessary dependencies.
+- Refer to any console logs for specific error messages to aid in debugging.