ms-agent

https://ms-agent-en.readthedocs.io

[{"step":"Define the task clearly. Use the prompt template to specify the goal (e.g., 'Automate X using Y') and include any input data or context (e.g., URLs, datasets, or constraints).","tip":"Be specific about the expected output format (e.g., report, code, or dataset). Include constraints like rate limits or tool preferences (e.g., 'Use Selenium for dynamic content')."},{"step":"Run the MS-Agent framework in your environment. Ensure Python and required libraries (e.g., `requests`, `selenium`, `beautifulsoup4`) are installed. Use a command like: `ms-agent --task \"[TASK_DESCRIPTION]\" --input [INPUT_DATA]`","tip":"For complex tasks, break the goal into smaller sub-tasks and run MS-Agent iteratively. Monitor logs for errors or warnings."},{"step":"Review the generated output. Verify the results against your expectations (e.g., check code functionality, data completeness). Use the output to refine the task or adjust parameters.","tip":"If the output is incomplete, add more context to the prompt (e.g., 'Include error handling for HTTP 500 errors') or re-run with adjusted parameters."},{"step":"Integrate the results into your workflow. For example, deploy the generated script, use the report in a dashboard, or feed the data into another tool.","tip":"Document the MS-Agent process for reproducibility. Save the prompt, output, and any generated files (e.g., code, reports) in a shared workspace."},{"step":"Iterate if needed. Use MS-Agent to refine the solution based on feedback or new requirements (e.g., 'Optimize the script for faster execution').","tip":"Track changes between iterations (e.g., using Git) to identify improvements or regressions."}]