{
    "class_name": "SequentialWorkFlowGraph",
    "goal": "Provide a concise answer to the question using relevant context. The answer must be straightforward and avoid unnecessary explanations.",
    "tasks": [
        {
            "name": "search_info",
            "description": "Extract the information and search.",
            "inputs": [
                {
                    "name": "question",
                    "type": "str",
                    "description": "The question that needs to be answered.",
                    "required": true
                }
            ],
            "outputs": [
                {
                    "name": "searching_output",
                    "type": "str",
                    "description": "The direct output after searching.",
                    "required": true
                }
            ],
            "prompt": "\"\"\"\nPlease search for and return relevant information related to {question}. Ensure the output is accurate, directly addresses the question, and provides necessary context. If the search yields insufficient or irrelevant information, clearly indicate this in the output. The output format should be in string. Additionally, validate that the information is correct and suitable for generating a concise answer before proceeding to the next task.\n\"\"\"",
            "prompt_template": {
                "class_name": "StringTemplate",
                "instruction": "Please search and return the related information in the question. The output format should be in string."
            },
            "system_prompt": "You are a helpful and highly intelligent assistant.",
            "parse_mode": "str",
            "parse_func": null,
            "parse_title": null,
            "tool_names": [
                "PertToolkit"
            ]
        },
        {
            "name": "generate_answer",
            "description": "Extract and formulate an answer from the given context.",
            "inputs": [
                {
                    "name": "question",
                    "type": "str",
                    "description": "The question that needs to be answered.",
                    "required": true
                },
                {
                    "name": "searching_output",
                    "type": "str",
                    "description": "Search information",
                    "required": true
                }
            ],
            "outputs": [
                {
                    "name": "answer",
                    "type": "str",
                    "description": "The direct answer to the question.",
                    "required": true
                }
            ],
            "prompt": "\"\"\"\nUse the context from `{searching_output}` to evaluate whether there is a significant change related to `{question}`. Clearly define \"significant change\" based on the relevant information provided in `{searching_output}` and any established criteria for significance. After your evaluation, validate your assessment against these criteria to ensure accuracy. Provide your final answer in a clear format (Final Answer: Yes or Final Answer: No).\n\"\"\"",
            "prompt_template": {
                "class_name": "StringTemplate",
                "instruction": "Use the context to determine the best answer to the question. Provide your final answer in a clear format (Yes or No) after reasoning."
            },
            "system_prompt": "You are a helpful and highly intelligent assistant.",
            "parse_mode": "xml",
            "parse_func": null,
            "parse_title": null,
            "tool_names": [
                "PertToolkit"
            ]
        },
        {
            "name": "validate_predictions7111",
            "description": "Task to validate_predictions7111. Takes searching_output as input. Produces validated_output as output.",
            "inputs": [
                {
                    "name": "searching_output",
                    "type": "str",
                    "description": "Input parameter searching_output for validate_predictions7111",
                    "required": false
                }
            ],
            "outputs": [
                {
                    "name": "validated_output",
                    "type": "str",
                    "description": "Output parameter validated_output from validate_predictions7111",
                    "required": true
                }
            ],
            "prompt": "```xml\n\"\"\"\nTo answer the question, first identify the relevant context that pertains to {question} and establish clear criteria for what constitutes a significant change in gene expression. Use this context to analyze the information gathered during the search, ensuring that you consider any potential discrepancies or systematic errors in the predictions. Validate the findings against established ground-truth solutions to ensure accuracy and check for any contradictory results before formulating the final answer. Clearly articulate your reasoning process in the 'thought' field, and provide the final answer in the 'answer' field. Ensure that your response is concise and directly addresses {question} without unnecessary elaboration. Format your output in xml format, such as <thought>{thought}</thought> and <answer>{answer}</answer>.\n\"\"\"\n```",
            "prompt_template": {
                "class_name": "StringTemplate",
                "instruction": "Think step by step to answer the question based on the question context. You should integrate context for answering. You should explain your thinking process in the 'thought' field, and provide the final answer in the 'answer' field.\nFormat your output in xml format, such as <thought>xxx</thought> and <answer>xxx</answer>."
            },
            "system_prompt": "You are a helpful and highly intelligent assistant.",
            "parse_mode": "xml",
            "parse_func": null,
            "parse_title": null,
            "tool_names": null
        }
    ]
}