{"id":27443008,"url":"https://github.com/fernicar/systemprobe","last_synced_at":"2026-04-30T13:34:16.264Z","repository":{"id":287167866,"uuid":"963565697","full_name":"fernicar/SystemProbe","owner":"fernicar","description":"an AI-powered desktop application designed to help users discover and optimize system prompts for Large Language Models (LLMs) without requiring complex fine-tuning or LoRA training.","archived":false,"fork":false,"pushed_at":"2025-04-10T16:42:47.000Z","size":374,"stargazers_count":1,"open_issues_count":0,"forks_count":0,"subscribers_count":1,"default_branch":"main","last_synced_at":"2025-04-15T01:18:08.165Z","etag":null,"topics":["app","application","chat-application","chatapp","chatbot","groq","groq-api","langchain","llm","pyside6","python","system-prompt","system-prompts","thereisnosource","tins"],"latest_commit_sha":null,"homepage":"","language":"Python","has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":"mit","status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/fernicar.png","metadata":{"files":{"readme":"README.md","changelog":null,"contributing":null,"funding":null,"license":"LICENSE","code_of_conduct":null,"threat_model":null,"audit":null,"citation":null,"codeowners":null,"security":null,"support":null,"governance":null,"roadmap":null,"authors":null,"dei":null,"publiccode":null,"codemeta":null}},"created_at":"2025-04-09T21:54:29.000Z","updated_at":"2025-04-10T16:42:51.000Z","dependencies_parsed_at":"2025-04-10T10:56:02.384Z","dependency_job_id":null,"html_url":"https://github.com/fernicar/SystemProbe","commit_stats":null,"previous_names":["fernicar/systemprobe"],"tags_count":0,"template":false,"template_full_name":null,"repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/fernicar%2FSystemProbe","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/fernicar%2FSystemProbe/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/fernicar%2FSystemProbe/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/fernicar%2FSystemProbe/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/fernicar","download_url":"https://codeload.github.com/fernicar/SystemProbe/tar.gz/refs/heads/main","host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":248986312,"owners_count":21194025,"icon_url":"https://github.com/github.png","version":null,"created_at":"2022-05-30T11:31:42.601Z","updated_at":"2022-07-04T15:15:14.044Z","host_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub","repositories_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories","repository_names_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repository_names","owners_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners"}},"keywords":["app","application","chat-application","chatapp","chatbot","groq","groq-api","langchain","llm","pyside6","python","system-prompt","system-prompts","thereisnosource","tins"],"created_at":"2025-04-15T01:18:21.599Z","updated_at":"2026-04-30T13:34:16.212Z","avatar_url":"https://github.com/fernicar.png","language":"Python","funding_links":[],"categories":[],"sub_categories":[],"readme":"# SystemProbe\n\n**SystemProbe** is an AI-powered desktop application designed to help users discover and optimize **system prompts** for Large Language Models (LLMs) without requiring complex fine-tuning or LoRA training. It guides users through an iterative process of refining system prompts to achieve desired outputs for dynamic inputs.\n\n## Features\n\n- **Dual LLM Workflow**: Uses two LLMs - one for testing prompts and another for refining them\n- **Iterative Refinement**: Step-by-step process to refine system prompts based on user feedback\n- **Visual Scoring**: Rate prompt effectiveness with an intuitive slider\n- **Custom Guidance**: Provide specific guidance to the refiner LLM\n- **Session Management**: Save and load your prompt optimization sessions\n- **Dark/Light Theme Support**: Choose your preferred visual theme\n- **Groq API Integration**: Leverages Groq's powerful LLM models\n\n## Usage\n\n1. **Start the application**:\n   ```\n   python main.py\n   ```\n   ![Step 0](https://github.com/fernicar/SystemProbe/blob/main/images/app_capture0.png)\n\n2. **Step 1: Define Inputs and Examples**\n   - Enter representative user inputs that your system prompt should handle\n   - Provide the ideal outputs for each input example\n   - Separate multiple examples with `---`\n   ![Step 1](https://github.com/fernicar/SystemProbe/blob/main/images/app_capture1.png)\n\n3. **Step 2: Set Initial System Prompt**\n   - Enter your starting system prompt\n   - This will be the baseline for refinement\n   ![Step 2](https://github.com/fernicar/SystemProbe/blob/main/images/app_capture2.png)\n\n4. **Step 3: Test Output and Score Results**\n   - The Tester LLM will generate output based on your system prompt\n   - Score the output from 1-10\n   - Provide specific feedback on what to improve\n   - Choose to refine or accept the prompt\n   ![Step 3](https://github.com/fernicar/SystemProbe/blob/main/images/app_capture3.png)\n\n5. **Step 4: Analyze and Refine**\n   - Review the Refiner LLM's analysis\n   - Add optional guidance for further refinement\n   - Click \"Refine Prompt\" to generate new suggestions\n   - Test the refined prompt or accept it as final\n   ![Step 4](https://github.com/fernicar/SystemProbe/blob/main/images/app_capture4.png)\n\n6. **Step 5: Final Optimized Prompt**\n   - Copy your optimized system prompt\n   - Save it to a file\n   - Start a new workflow if needed\n   ![Step 5](https://github.com/fernicar/SystemProbe/blob/main/images/app_capture5.png)\n\n## Installation\n\n1. Clone the repository:\n   ```\n   git clone https://github.com/fernicar/SystemProbe.git\n   cd SystemProbe\n   ```\n\n2. Create and activate a virtual environment:\n   ```\n   python -m venv .venv\n   .venv\\Scripts\\activate\n   ```\n\n3. Install dependencies:\n   ```\n   pip install -r requirements.txt\n   ```\n\n4. Set up your Groq free API key:\n   - Create a `.env` file in the project root with:\n     ```\n     GROQ_API_KEY='your_groq_api_key_here'\n     ```\n   - Or enter it in the application settings\n\n## Configuration\n\n- **API Key**: Set your Groq API key in Settings\n- **Theme**: Choose between Dark and Light themes\n- **LLM Model**: Select from available Groq models\n- **Model Updates**: Toggle automatic model list updates\n\n## Technologies Used\n\n- **Python**: Core programming language\n- **PySide6**: Qt-based GUI framework\n- **Langchain**: Framework for LLM application development\n- **Groq API**: High-performance LLM provider\n- **QThread Workers**: For non-blocking LLM operations\n\n## License\n\nThis project is licensed under the MIT License - see the [LICENSE](https://github.com/fernicar/SystemProbe/blob/main/LICENSE) file for details.\n\n## Acknowledgments\n\n*   Special thanks to ScuffedEpoch for the [TINS](https://github.com/ScuffedEpoch/TINS) methodology and the initial example.\n*   Thanks to the free tier AI assistant for its initial contribution to the project.\n*   Gratitude to the Groq team for their API and support.\n*   Thanks to the Langchain and PySide6 communities for their respective libraries and documentation.\n*   Augment extension for VS Code\n*   Tested LLM Gemini2.5pro (free tier beta testing) from Google AI Studio\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Ffernicar%2Fsystemprobe","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Ffernicar%2Fsystemprobe","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Ffernicar%2Fsystemprobe/lists"}