app: description: The chatboot will be able to answer only about our documents context icon: 🤖 icon_background: '#FFEAD5' icon_type: emoji mode: advanced-chat name: 'Exercise 4: Building an advanced RAG' use_icon_as_answer_icon: false dependencies: - current_identifier: null type: marketplace value: marketplace_plugin_unique_identifier: langgenius/azure_openai:0.0.69@c744f6bf60699c296307a4420094a7f88b4f6e7fdf353b67b2b88ddb455074b2 version: null kind: app version: 0.7.0 workflow: conversation_variables: [] environment_variables: [] features: file_upload: allowed_file_extensions: - .JPG - .JPEG - .PNG - .GIF - .WEBP - .SVG allowed_file_types: - image allowed_file_upload_methods: - local_file - remote_url enabled: false fileUploadConfig: audio_file_size_limit: 50 batch_count_limit: 5 file_size_limit: 15 image_file_size_limit: 10 video_file_size_limit: 100 workflow_file_upload_limit: 10 image: enabled: false number_limits: 3 transfer_methods: - local_file - remote_url number_limits: 3 opening_statement: "Hi there! I'm your AI-powered testing assistant. Here's what\ \ I can help you with:\n\n • Ask me about specific Jira issues (e.g., \"\ What does REST-433 fix?\")\n • Request summaries or comparisons for multiple\ \ issues\n • Get Gherkin-style test cases based on requirements or bugs\n\ \ • Ask questions about your project’s scope, documentation, or contributors\n\ \n Just type your question below and I’ll take it from there!" retriever_resource: enabled: true sensitive_word_avoidance: enabled: false speech_to_text: enabled: false suggested_questions: - What is this project about? - How to test issue REST-266? - Create test scenarios for issue REST-320 - Find issues related to Jira API suggested_questions_after_answer: enabled: true text_to_speech: enabled: false language: '' voice: '' graph: edges: - data: isInIteration: false isInLoop: false sourceType: llm targetType: answer id: llm-answer selected: false source: llm sourceHandle: source target: answer targetHandle: target type: custom zIndex: 0 - data: isInIteration: false isInLoop: false sourceType: knowledge-retrieval targetType: llm id: 1748687775954-source-llm-target selected: false source: '1748687775954' sourceHandle: source target: llm targetHandle: target type: custom zIndex: 0 - data: isInIteration: false isInLoop: false sourceType: parameter-extractor targetType: knowledge-retrieval id: 1748719005808-source-1748687775954-target source: '1748719005808' sourceHandle: source target: '1748687775954' targetHandle: target type: custom zIndex: 0 - data: isInIteration: false isInLoop: false sourceType: start targetType: question-classifier id: 1748687003240-source-1748756681684-target source: '1748687003240' sourceHandle: source target: '1748756681684' targetHandle: target type: custom zIndex: 0 - data: isInIteration: false isInLoop: false sourceType: question-classifier targetType: parameter-extractor id: 1748756681684-1-1748719005808-target source: '1748756681684' sourceHandle: '1' target: '1748719005808' targetHandle: target type: custom zIndex: 0 - data: isInIteration: false isInLoop: false sourceType: question-classifier targetType: knowledge-retrieval id: 1748756681684-1748767162845-1748767429892-target source: '1748756681684' sourceHandle: '1748767162845' target: '1748767429892' targetHandle: target type: custom zIndex: 0 - data: isInIteration: false isInLoop: false sourceType: knowledge-retrieval targetType: llm id: 1748767429892-source-1748767447593-target source: '1748767429892' sourceHandle: source target: '1748767447593' targetHandle: target type: custom zIndex: 0 - data: isInIteration: false isInLoop: false sourceType: llm targetType: answer id: 1748767447593-source-1748767451380-target source: '1748767447593' sourceHandle: source target: '1748767451380' targetHandle: target type: custom zIndex: 0 - data: isInIteration: false isInLoop: false sourceType: question-classifier targetType: parameter-extractor id: 1748756681684-2-1748767570289-target source: '1748756681684' sourceHandle: '2' target: '1748767570289' targetHandle: target type: custom zIndex: 0 - data: isInIteration: false isInLoop: false sourceType: llm targetType: answer id: 1748783029512-source-1748783041984-target source: '1748783029512' sourceHandle: source target: '1748783041984' targetHandle: target type: custom zIndex: 0 - data: isInIteration: false isInLoop: false sourceType: parameter-extractor targetType: iteration id: 1748767570289-source-1748783526414-target source: '1748767570289' sourceHandle: source target: '1748783526414' targetHandle: target type: custom zIndex: 0 - data: isInIteration: true isInLoop: false iteration_id: '1748783526414' sourceType: iteration-start targetType: knowledge-retrieval id: 1748783526414start-source-1748783571565-target source: 1748783526414start sourceHandle: source target: '1748783571565' targetHandle: target type: custom zIndex: 1001 - data: isInIteration: false isInLoop: false sourceType: question-classifier targetType: knowledge-retrieval id: 1748756681684-1748767109717-1748784602400-target source: '1748756681684' sourceHandle: '1748767109717' target: '1748784602400' targetHandle: target type: custom zIndex: 0 - data: isInIteration: false isInLoop: false sourceType: knowledge-retrieval targetType: llm id: 1748784602400-source-1748784677999-target source: '1748784602400' sourceHandle: source target: '1748784677999' targetHandle: target type: custom zIndex: 0 - data: isInIteration: false isInLoop: false sourceType: llm targetType: answer id: 1748784677999-source-1748784661836-target source: '1748784677999' sourceHandle: source target: '1748784661836' targetHandle: target type: custom zIndex: 0 - data: isInIteration: false isInLoop: false sourceType: question-classifier targetType: parameter-extractor id: 1748756681684-1748774938291-17487850509710-target source: '1748756681684' sourceHandle: '1748774938291' target: '17487850509710' targetHandle: target type: custom zIndex: 0 - data: isInIteration: true isInLoop: false iteration_id: '17487852863990' sourceType: iteration-start targetType: knowledge-retrieval id: 1748785286399start-source-1748785286399017487852864000-target source: 1748785286399start sourceHandle: source target: '1748785286399017487852864000' targetHandle: target type: custom zIndex: 1001 - data: isInIteration: false isInLoop: false sourceType: parameter-extractor targetType: if-else id: 17487850509710-source-1748785479296-target source: '17487850509710' sourceHandle: source target: '1748785479296' targetHandle: target type: custom zIndex: 0 - data: isInIteration: false isInLoop: false sourceType: if-else targetType: iteration id: 1748785479296-true-17487852863990-target source: '1748785479296' sourceHandle: 'true' target: '17487852863990' targetHandle: target type: custom zIndex: 0 - data: isInIteration: false isInLoop: false sourceType: if-else targetType: llm id: 1748785479296-f2be164b-f0ce-4db6-9a7b-69786c02a988-1748785532843-target source: '1748785479296' sourceHandle: f2be164b-f0ce-4db6-9a7b-69786c02a988 target: '1748785532843' targetHandle: target type: custom zIndex: 0 - data: isInIteration: false isInLoop: false sourceType: llm targetType: answer id: 1748785557565-source-1748785836978-target source: '1748785557565' sourceHandle: source target: '1748785836978' targetHandle: target type: custom zIndex: 0 - data: isInIteration: false isInLoop: false sourceType: llm targetType: answer id: 1748785532843-source-1748786521492-target source: '1748785532843' sourceHandle: source target: '1748786521492' targetHandle: target type: custom zIndex: 0 - data: isInIteration: false isInLoop: false sourceType: question-classifier targetType: llm id: 1748756681684-1748785172818-1748786653855-target source: '1748756681684' sourceHandle: '1748785172818' target: '1748786653855' targetHandle: target type: custom zIndex: 0 - data: isInIteration: false isInLoop: false sourceType: llm targetType: answer id: 1748786653855-source-1748785149607-target source: '1748786653855' sourceHandle: source target: '1748785149607' targetHandle: target type: custom zIndex: 0 - data: isInIteration: false isInLoop: false sourceType: iteration targetType: code id: 1748783526414-source-1762727584347-target source: '1748783526414' sourceHandle: source target: '1762727584347' targetHandle: target type: custom zIndex: 0 - data: isInIteration: false isInLoop: false sourceType: code targetType: llm id: 1762727584347-source-1748783029512-target source: '1762727584347' sourceHandle: source target: '1748783029512' targetHandle: target type: custom zIndex: 0 - data: isInIteration: false isInLoop: false sourceType: iteration targetType: code id: 17487852863990-source-1762728060345-target source: '17487852863990' sourceHandle: source target: '1762728060345' targetHandle: target type: custom zIndex: 0 - data: isInIteration: false isInLoop: false sourceType: code targetType: llm id: 1762728060345-source-1748785557565-target source: '1762728060345' sourceHandle: source target: '1748785557565' targetHandle: target type: custom zIndex: 0 nodes: - data: desc: '' selected: false title: Start type: start variables: [] height: 73 id: '1748687003240' position: x: -1083.666487928184 y: 288.65445462792803 positionAbsolute: x: -1083.666487928184 y: 288.65445462792803 selected: false sourcePosition: right targetPosition: left type: custom width: 242 zIndex: 0 - data: context: enabled: true variable_selector: - '1748687775954' - result desc: '' memory: query_prompt_template: '{{#sys.query#}}' role_prefix: assistant: '' user: '' window: enabled: false size: 10 model: completion_params: {} mode: chat name: gpt-35-turbo-16k provider: langgenius/azure_openai/azure_openai prompt_template: - id: 5c286fbf-33f9-4150-8c64-696789577558 role: system text: "You are a software testing assistant helping engineers understand\ \ Jira issues, project documentation, and related work.\n\nOnly answer\ \ questions using the information provided in the context below.\nIf the\ \ context does not contain relevant information to answer the user's question,\ \ clearly say so.\n\nContext documents are formatted like this:\n\n- Jira\ \ issues follow this structure:\n Jira Issue: \n Project: \n Type: \n Status: \n\ \ Assignee: \n Created: \n Updated: \n\ \n Summary: \n\n Description:\n \n\ \n- Summaries and technical documentation may also include:\n Summary,\ \ Contributors, Assignees, Reporters, Issue Count, Type.\n\nUse this format\ \ to extract and organize key information when answering.\n\nYour reply\ \ should:\n- Be concise and professional\n- Highlight relevant fields\ \ (e.g., Summary, Type, Assignee)\n- Mention related issues if they exist\ \ in the context\n- Focus on the implications for software testing when\ \ possible\n- Never answer using outside knowledge or speculation\n\n\ If the question is outside the scope of the documents (e.g. “Why is the\ \ sky blue?”), respond with:\n> I'm sorry, I can't find relevant information\ \ in the project documentation to answer that.\n\nOnly answer using the\ \ provided context.\n\nOnly answer using the provided context.\n\nIf the\ \ context does NOT clearly mention the Jira issue referenced in the user's\ \ question (e.g. REST-XXX, issue XXX, XXX, Jira Issue XXX), then respond\ \ with:\n\n> I’m sorry, I can’t find details about that specific issue\ \ in the project documentation.\n\n\n\n\n\nContext:\n{{#context#}}\n\n\ User question:\n{{#sys.query#}}\n\nReminder: If the Jira issue mentioned\ \ above is not present in the context chunks, do not guess or fabricate\ \ the answer.\nExamples:\n\nQ: What does REST-433 fix?\nContext includes\ \ \"Jira Issue: REST-433\"\n✅ Answer with issue details.\n\nQ: What does\ \ REST-433 fix?\nContext does NOT include REST-433\n❌ Do not guess. Respond\ \ with: \"I’m sorry, I can’t find details about that specific issue...\"\ \n\n" selected: false title: LLM type: llm variables: [] vision: enabled: false height: 88 id: llm position: x: 456.2543663027028 y: 33.57027864430731 positionAbsolute: x: 456.2543663027028 y: 33.57027864430731 selected: false sourcePosition: right targetPosition: left type: custom width: 242 zIndex: 0 - data: answer: '{{#llm.text#}}' desc: '' selected: false title: Answer Specific Issue Queries type: answer variables: [] height: 103 id: answer position: x: 792.566799557453 y: 33.57027864430731 positionAbsolute: x: 792.566799557453 y: 33.57027864430731 selected: false sourcePosition: right targetPosition: left type: custom width: 242 zIndex: 0 - data: dataset_ids: - 6cKbMQM8KeUisYYXzRoSVTZwEObucVVwsbtpxzAyNJ6iSCNdcrCcOfXhQnGIVVI2 desc: '' metadata_filtering_conditions: conditions: - comparison_operator: is id: 2cc78fcc-d7fd-49e1-a66d-e71af312b069 name: issue_key value: '{{#1748719005808.issue_key#}}' logical_operator: and metadata_filtering_mode: manual metadata_model_config: completion_params: temperature: 0.7 mode: chat name: gpt-35-turbo-16k provider: langgenius/azure_openai/azure_openai multiple_retrieval_config: reranking_enable: false reranking_mode: weighted_score reranking_model: model: '' provider: '' score_threshold: 0.31 top_k: 10 weights: keyword_setting: keyword_weight: 0 vector_setting: embedding_model_name: text-embedding-3-large embedding_provider_name: langgenius/azure_openai/azure_openai vector_weight: 1 weight_type: customized query_variable_selector: - '1748687003240' - sys.query retrieval_mode: multiple selected: false title: KR - Single Issue type: knowledge-retrieval height: 90 id: '1748687775954' position: x: 89.0778811944802 y: 33.57027864430731 positionAbsolute: x: 89.0778811944802 y: 33.57027864430731 selected: false sourcePosition: right targetPosition: left type: custom width: 242 zIndex: 0 - data: desc: '' instruction: "You are an intelligent assistant specialized in understanding\ \ user requests related to Jira issues and test case management. Your primary\ \ goal is to analyze the user's {{#sys.query#}} and extract one piece of\ \ information in a precise JSON format: \n1. **`issue_key`**: A Jira issue\ \ key, if explicitly mentioned or strongly implied. \n\n\n --- **Detailed\ \ Instructions for `issue_key` Extraction:** * An issue key follows the\ \ format: `[ONE_OR_MORE_UPPERCASE_LETTERS]-[ONE_OR_MORE_DIGITS]`. * Examples:\ \ `PROJ-123`, `BUG-42`, `INC-005`, `SUPPORT-789`. * Look for the issue key\ \ appearing after phrases like 'related to issue', 'ticket:', 'issue:',\ \ 'for ticket', 'for issue', or within the user's general statement. * If\ \ no string matches the exact 'PROJNAME-XXX' format (e.g., '123-PROJ', 'PROJ123',\ \ 'PROJ-ABC'), correct and give appropiate format and return this `issue_key`.\ \ \n/\n" model: completion_params: temperature: 0.7 mode: chat name: gpt-35-turbo-16k provider: langgenius/azure_openai/azure_openai parameters: - description: 'An issue key typically consists of an uppercase project prefix, followed by an hyphen, and then one or more digits. For example: JIRA-123, PROJ-45, BUG-007.' name: issue_key required: true type: string query: - sys - query reasoning_mode: prompt selected: false title: Single Issue Key Extractor type: parameter-extractor variables: [] vision: enabled: false height: 88 id: '1748719005808' position: x: -284.44013680327953 y: 33.57027864430731 positionAbsolute: x: -284.44013680327953 y: 33.57027864430731 selected: false sourcePosition: right targetPosition: left type: custom width: 242 zIndex: 0 - data: classes: - id: '1' name: ' specific_issue' - id: '2' name: list_issues - id: '1748767109717' name: general_project - id: '1748767162845' name: test_related - id: '1748774938291' name: test_generation - id: '1748785172818' name: unrelated_or_invalid desc: '' instruction: " You are a smart assistant helping with software testing,\ \ Jira issues, and documentation.\n\n Your task is to classify the user's\ \ query into ONE of the following categories:\n\n - `specific_issue`:\ \ The user is asking about one particular Jira issue.\n > Example:\ \ \"What does REST-433 fix?\" or \"Who closed issue BUG-123?\"\n\n -\ \ `list_issues`: The user is referencing multiple **explicit issue keys**\ \ (e.g. REST-123, BUG-456). This category applies **only** when two or more\ \ issue keys are clearly mentioned.\n > Example: \"How are REST-433\ \ and REST-426 related?\"\n\n - `general_project`: The question relates\ \ to overall project scope, architecture, documentation, contributors, or\ \ themes of work. Includes filtered requests like \"List bugs related to\ \ the Jira plugin\" when no issue keys are named.\n > Example: \"What\ \ is this project about?\" or \"List some bugs related to Jira plugin\"\n\ \n - `test_related`: The user is asking about testing activities, strategies,\ \ or coverage.\n > Example: \"What is the test plan for this feature?\"\ \ or \"Which tests cover the login module?\"\n\n - `test_generation`:\ \ The user is asking to generate test cases. This includes:\n • questions\ \ starting with \"write\", \"generate\", or \"create test\"\n • generation\ \ requests from bugs, requirements, or user text\n > Example: \"Create\ \ test cases for password reset\" or \"Generate tests for REST-433 and REST-434\"\ \n\n - `unrelated_or_invalid`: The query is not related to testing, documentation,\ \ or issue analysis.\n > Example: \"Write me a poem\" or \"What's the\ \ weather today?\"\n\n ⚠️ Always return ONE label from:\n `specific_issue`,\ \ `list_issues`, `general_project`, `test_related`, `test_generation`, `unrelated_or_invalid`\n\ \n Do NOT explain your answer.\n\n ---\n User query:\n {{#sys.query#}}\n" instructions: '' model: completion_params: temperature: 0.7 mode: chat name: gpt-35-turbo-16k provider: langgenius/azure_openai/azure_openai query_variable_selector: - '1748687003240' - sys.query selected: false title: Question Classifier topics: [] type: question-classifier vision: enabled: false height: 346 id: '1748756681684' position: x: -694.8206093064899 y: 288.65445462792803 positionAbsolute: x: -694.8206093064899 y: 288.65445462792803 selected: false sourcePosition: right targetPosition: left type: custom width: 242 zIndex: 0 - data: dataset_ids: - 6cKbMQM8KeUisYYXzRoSVTZwEObucVVwsbtpxzAyNJ6iSCNdcrCcOfXhQnGIVVI2 desc: '' multiple_retrieval_config: reranking_enable: false reranking_mode: weighted_score top_k: 4 weights: keyword_setting: keyword_weight: 0 vector_setting: embedding_model_name: text-embedding-3-large embedding_provider_name: langgenius/azure_openai/azure_openai vector_weight: 1 weight_type: customized query_variable_selector: - '1748687003240' - sys.query retrieval_mode: multiple selected: false title: KR General Knowledge type: knowledge-retrieval height: 90 id: '1748767429892' position: x: -284.44013680327953 y: 610.4803953947226 positionAbsolute: x: -284.44013680327953 y: 610.4803953947226 selected: false sourcePosition: right targetPosition: left type: custom width: 242 zIndex: 0 - data: context: enabled: false variable_selector: [] desc: '' model: completion_params: temperature: 0.7 mode: chat name: gpt-35-turbo-16k provider: langgenius/azure_openai/azure_openai prompt_template: - id: 8c741de7-7353-402c-bfcb-a1e6b6798e35 role: system text: 'You are a software testing assistant helping engineers understand project-level information, contributors, documentation, and testing scope. Only answer questions using the information provided in the context below. If the context does not contain relevant information to answer the user''s question, clearly say so. Context documents may include: - Project summaries, descriptions, and objectives - Contributor and assignee lists - Component or module overviews - Documentation of testing scope or strategies - Aggregated metrics such as issue count by type or status Do not answer about specific Jira issues unless they are mentioned as part of a broader project summary. Your reply should: - Be concise and professional - Highlight key facts about the project (e.g., modules, teams, goals, test scope) - Mention totals or distributions (e.g., "There are 12 open bugs", "Main modules are X, Y") - Focus on information relevant to project-level understanding or software testing readiness - Never answer using outside knowledge or speculation If the user’s question is not answered in the context, respond with: > I''m sorry, I can''t find relevant project-level information in the documentation to answer that. Only answer using the provided context. --- Context: {{#context#}} User question: {{#sys.query#}} Examples: Q: What does the project include? Context includes a project summary and modules ✅ Answer with module names and summary. Q: What modules are included? Context does NOT mention modules ❌ Do not guess. Respond: "I can''t find relevant project-level information..." Q: What’s the testing strategy for this release? Context includes QA section ✅ Answer with scope or strategy Q: What’s the testing strategy for this release? Context does NOT include any testing info ❌ Do not guess.' selected: false title: LLM 2 type: llm variables: [] vision: enabled: false height: 88 id: '1748767447593' position: x: 89.0778811944802 y: 610.4803953947226 positionAbsolute: x: 89.0778811944802 y: 610.4803953947226 selected: false sourcePosition: right targetPosition: left type: custom width: 242 zIndex: 0 - data: answer: '{{#1748767447593.text#}}' desc: '' selected: false title: Answer 2 type: answer variables: [] height: 103 id: '1748767451380' position: x: 456.2543663027028 y: 610.4803953947226 positionAbsolute: x: 456.2543663027028 y: 610.4803953947226 selected: false sourcePosition: right targetPosition: left type: custom width: 242 zIndex: 0 - data: desc: '' instruction: "You are an intelligent assistant specialized in understanding\ \ user requests related to Jira issues and test case management. Your primary\ \ goal is to analyze the user's {{#sys.query#}} and extract all mentioned\ \ Jira issue keys in a precise JSON format:\n\n1. **`issue_keys`**: A list\ \ of Jira issue keys explicitly mentioned or strongly implied in the user's\ \ query.\n\n---\n\n**Detailed Instructions for `issue_keys` Extraction:**\n\ \n- An issue key follows the format: `[ONE_OR_MORE_UPPERCASE_LETTERS]-[ONE_OR_MORE_DIGITS]`.\n\ \ - Examples: `PROJ-123`, `BUG-42`, `INC-005`, `SUPPORT-789`.\n- Look for\ \ issue keys appearing after phrases like:\n - 'related to issues'\n -\ \ 'tickets:'\n - 'issues:'\n - 'for tickets'\n - 'for issues'\n - or\ \ within the user's general statement.\n- If a string closely resembles\ \ an issue key but deviates slightly (e.g., '123-PROJ', 'PROJ123', 'PROJ-ABC'),\ \ correct it to the appropriate format and include it in the `issue_keys`\ \ list.\n- If no valid issue keys are found, return an empty list.\n\n---\n\ \n**Output Format:**\n\n```json\n{\n \"issue_keys\": [\"PROJ-123\", \"\ BUG-42\"]\n}" model: completion_params: temperature: 0.7 mode: chat name: gpt-35-turbo-16k provider: langgenius/azure_openai/azure_openai parameters: - description: 'A coma separated list of issue keys that the user is interested in. An issue key typically consists of an uppercase project prefix, followed by an hyphen, and then one or more digits. For example: JIRA-123, PROJ-45, BUG-007.' name: issue_keys required: true type: array[string] query: - sys - query reasoning_mode: prompt selected: false title: Issue Keys Extractor type: parameter-extractor variables: [] vision: configs: detail: high variable_selector: - sys - files enabled: false height: 88 id: '1748767570289' position: x: -284.44013680327953 y: 211.5747334168887 positionAbsolute: x: -284.44013680327953 y: 211.5747334168887 selected: false sourcePosition: right targetPosition: left type: custom width: 242 zIndex: 0 - data: context: enabled: true variable_selector: - '1762727584347' - flat_context desc: '' model: completion_params: temperature: 0.7 mode: chat name: gpt-35-turbo-16k provider: langgenius/azure_openai/azure_openai prompt_config: jinja2_variables: [] prompt_template: - edition_type: basic id: 41d8a4b2-979b-4166-a924-d9b9e7b724da role: system text: "You are a software testing assistant helping engineers understand\ \ Jira issues, project documentation, and related work.\n\nOnly answer\ \ questions using the information provided in the context below.\nIf the\ \ context does not contain relevant information to answer the user's question,\ \ clearly say so.\n\nContext documents are formatted like this:\n\n- Jira\ \ issues follow this structure:\n Jira Issue: \n Project: \n Type: \n Status: \n\ \ Assignee: \n Created: \n Updated: \n\ \n Summary: \n\n Description:\n \n\ \n- Summaries and technical documentation may also include:\n Summary,\ \ Contributors, Assignees, Reporters, Issue Count, Type.\n\nUse this format\ \ to extract and organize key information when answering.\n\nYour reply\ \ should:\n- Be concise and professional\n- Highlight relevant fields\ \ (e.g., Summary, Type, Assignee)\n- Mention related issues if they exist\ \ in the context\n- Focus on the implications for software testing when\ \ possible\n- Never answer using outside knowledge or speculation\n\n\ If the question is outside the scope of the documents (e.g. “Why is the\ \ sky blue?”), respond with:\n> I'm sorry, I can't find relevant information\ \ in the project documentation to answer that.\n\nOnly answer using the\ \ provided context.\n\nOnly answer using the provided context.\n\nIf the\ \ context does NOT clearly mention the Jira issue referenced in the user's\ \ question (e.g. REST-XXX, issue XXX, XXX, Jira Issue XXX), then respond\ \ with:\n\n> I’m sorry, I can’t find details about that specific issue\ \ in the project documentation.\n\n\nContext:\n\n\nUser question:\n{{#sys.query#}}\n\ \nReminder: If the Jira issue mentioned above is not present in the context\ \ chunks, do not guess or fabricate the answer.\nExamples:\n\nQ: What\ \ does REST-433 fix?\nContext includes \"Jira Issue: REST-433\"\n✅ Answer\ \ with issue details.\n\nQ: What does REST-433 fix?\nContext does NOT\ \ include REST-433\n❌ Do not guess. Respond with: \"I’m sorry, I can’t\ \ find details about that specific issue...\"\n\n{{#context#}}\n\nBatch\ \ completion status: {{#1762727584347.batch_status#}}\nIf this status\ \ is partial or empty, state it clearly before presenting any successful\ \ results." selected: false title: LLM 3 type: llm variables: [] vision: enabled: false height: 88 id: '1748783029512' position: x: 1092.5667995574531 y: 211.5747334168887 positionAbsolute: x: 1092.5667995574531 y: 211.5747334168887 selected: false sourcePosition: right targetPosition: left type: custom width: 242 zIndex: 0 - data: answer: '{{#1748783029512.text#}}' desc: '' selected: false title: Answer 3 type: answer variables: [] height: 103 id: '1748783041984' position: x: 1436.1121280862612 y: 211.5747334168887 positionAbsolute: x: 1436.1121280862612 y: 211.5747334168887 selected: false sourcePosition: right targetPosition: left type: custom width: 242 zIndex: 0 - data: desc: '' error_handle_mode: remove-abnormal-output height: 187 is_parallel: true iterator_input_type: array[string] iterator_selector: - '1748767570289' - issue_keys output_selector: - '1748783571565' - result output_type: array[object] parallel_nums: 10 selected: false start_node_id: 1748783526414start title: Iteration type: iteration width: 618 height: 187 id: '1748783526414' position: x: 89.0778811944802 y: 211.5747334168887 positionAbsolute: x: 89.0778811944802 y: 211.5747334168887 selected: false sourcePosition: right targetPosition: left type: custom width: 618 zIndex: 0 - data: desc: '' isInIteration: true selected: false title: '' type: iteration-start draggable: false height: 48 id: 1748783526414start parentId: '1748783526414' position: x: 24 y: 68 positionAbsolute: x: 113.0778811944802 y: 279.5747334168887 selectable: false sourcePosition: right targetPosition: left type: custom-iteration-start width: 44 zIndex: 1001 - data: dataset_ids: - 6cKbMQM8KeUisYYXzRoSVTZwEObucVVwsbtpxzAyNJ6iSCNdcrCcOfXhQnGIVVI2 desc: '' isInIteration: true isInLoop: false iteration_id: '1748783526414' metadata_filtering_conditions: conditions: - comparison_operator: is id: 2cc78fcc-d7fd-49e1-a66d-e71af312b069 name: issue_key value: '{{#1748783526414.item#}}' logical_operator: and metadata_filtering_mode: manual multiple_retrieval_config: reranking_enable: false reranking_mode: weighted_score reranking_model: model: '' provider: '' score_threshold: null top_k: 10 weights: keyword_setting: keyword_weight: 0 vector_setting: embedding_model_name: text-embedding-3-large embedding_provider_name: langgenius/azure_openai/azure_openai vector_weight: 1 weight_type: customized query_variable_selector: - '1748783526414' - item retrieval_mode: multiple selected: false title: KR - Multiple Issues type: knowledge-retrieval height: 90 id: '1748783571565' parentId: '1748783526414' position: x: 213.39028285937735 y: 65 positionAbsolute: x: 302.46816405385755 y: 276.5747334168887 selected: false sourcePosition: right targetPosition: left type: custom width: 242 zIndex: 1001 - data: dataset_ids: - 6cKbMQM8KeUisYYXzRoSVTZwEObucVVwsbtpxzAyNJ6iSCNdcrCcOfXhQnGIVVI2 desc: '' multiple_retrieval_config: reranking_enable: false reranking_mode: weighted_score reranking_model: model: '' provider: '' score_threshold: null top_k: 10 weights: keyword_setting: keyword_weight: 0 vector_setting: embedding_model_name: text-embedding-3-large embedding_provider_name: langgenius/azure_openai/azure_openai vector_weight: 1 weight_type: customized query_variable_selector: - sys - query retrieval_mode: multiple selected: false title: 'KR General Knowledge ' type: knowledge-retrieval height: 90 id: '1748784602400' position: x: -284.44013680327953 y: 464.40811576196575 positionAbsolute: x: -284.44013680327953 y: 464.40811576196575 selected: false sourcePosition: right targetPosition: left type: custom width: 242 zIndex: 0 - data: answer: '{{#1748784677999.text#}}/' desc: '' selected: false title: Answer 4 type: answer variables: [] height: 103 id: '1748784661836' position: x: 456.2543663027028 y: 464.40811576196575 positionAbsolute: x: 456.2543663027028 y: 464.40811576196575 selected: false sourcePosition: right targetPosition: left type: custom width: 242 zIndex: 0 - data: context: enabled: true variable_selector: - '1748784602400' - result desc: '' model: completion_params: temperature: 0.7 mode: chat name: gpt-35-turbo-16k provider: langgenius/azure_openai/azure_openai prompt_template: - id: b3c07274-69ce-4bb7-843b-e6db08c42df1 role: system text: 'You are a software testing assistant helping engineers understand project, product, and company context. Only answer questions using the retrieved context provided below. Do not make assumptions or use outside knowledge. Your goal is to synthesize information across the provided context and answer the user''s question as fully as possible. If relevant content is spread across multiple context blocks, combine them into a cohesive answer. Context documents may include: - Project summaries or goals - Technical overviews or API scopes - Lists of contributors or reporters - Fixes, optimizations, and technical constraints Your response should: - Be concise and professional - Highlight specific facts from the context - Focus on what the project is about and why it matters - Relate the information to software testing implications if relevant If no relevant information can be found in the context, respond with: > I’m sorry, I can’t find relevant information in the documentation to answer that. --- Context: {{#context#}} User question: {{#sys.query#}} ' selected: false structured_output_enabled: false title: LLM 4 type: llm variables: [] vision: enabled: false height: 88 id: '1748784677999' position: x: 89.0778811944802 y: 464.40811576196575 positionAbsolute: x: 89.0778811944802 y: 464.40811576196575 selected: false sourcePosition: right targetPosition: left type: custom width: 242 zIndex: 0 - data: desc: '' instruction: "You are an intelligent assistant specialized in understanding\ \ user requests related to Jira issues and test case management. Your primary\ \ goal is to analyze the user's {{#sys.query#}} and extract all mentioned\ \ Jira issue keys in a precise JSON format:\n\n1. **`issue_keys`**: A list\ \ of Jira issue keys explicitly mentioned or strongly implied in the user's\ \ query.\n\n---\n\n**Detailed Instructions for `issue_keys` Extraction:**\n\ \n- An issue key follows the format: `[ONE_OR_MORE_UPPERCASE_LETTERS]-[ONE_OR_MORE_DIGITS]`.\n\ \ - Examples: `PROJ-123`, `BUG-42`, `INC-005`, `SUPPORT-789`.\n- Look for\ \ issue keys appearing after phrases like:\n - 'related to issues'\n -\ \ 'tickets:'\n - 'issues:'\n - 'for tickets'\n - 'for issues'\n - or\ \ within the user's general statement.\n- If a string closely resembles\ \ an issue key but deviates slightly (e.g., '123-PROJ', 'PROJ123', 'PROJ-ABC'),\ \ correct it to the appropriate format and include it in the `issue_keys`\ \ list.\n- If no valid issue keys are found, return an empty list.\n\n---\n\ \n**Output Format:**\n\n```json\n{\n \"issue_keys\": [\"PROJ-123\", \"\ BUG-42\"]\n}" model: completion_params: temperature: 0.7 mode: chat name: gpt-35-turbo-16k provider: langgenius/azure_openai/azure_openai parameters: - description: 'A coma separated list of issue keys that the user is interested in. An issue key typically consists of an uppercase project prefix, followed by an hyphen, and then one or more digits. For example: JIRA-123, PROJ-45, BUG-007.' name: issue_keys required: true type: array[string] query: - sys - query reasoning_mode: prompt selected: false title: Issue Keys Extractor (1) type: parameter-extractor variables: [] vision: configs: detail: high variable_selector: - sys - files enabled: false height: 88 id: '17487850509710' position: x: -284.44013680327953 y: 754.9250298061089 positionAbsolute: x: -284.44013680327953 y: 754.9250298061089 selected: false sourcePosition: right targetPosition: left type: custom width: 242 zIndex: 0 - data: answer: '{{#1748786653855.text#}}' desc: '' selected: false title: Answer 5 type: answer variables: [] height: 103 id: '1748785149607' position: x: 89.0778811944802 y: 1011.2915512785755 positionAbsolute: x: 89.0778811944802 y: 1011.2915512785755 selected: false sourcePosition: right targetPosition: left type: custom width: 242 zIndex: 0 - data: desc: '' error_handle_mode: remove-abnormal-output height: 187 is_parallel: true iterator_selector: - '17487850509710' - issue_keys output_selector: - '1748785286399017487852864000' - result output_type: array[object] parallel_nums: 10 selected: false start_node_id: 1748785286399start title: Iteration (1) type: iteration width: 618 height: 187 id: '17487852863990' position: x: 456.2543663027028 y: 743.564737158242 positionAbsolute: x: 456.2543663027028 y: 743.564737158242 selected: false sourcePosition: right targetPosition: left type: custom width: 618 zIndex: 0 - data: dataset_ids: - 6cKbMQM8KeUisYYXzRoSVTZwEObucVVwsbtpxzAyNJ6iSCNdcrCcOfXhQnGIVVI2 desc: '' isInIteration: true isInLoop: false iteration_id: '17487852863990' metadata_filtering_conditions: conditions: - comparison_operator: is id: 2cc78fcc-d7fd-49e1-a66d-e71af312b069 name: issue_key value: '{{#17487852863990.item#}}' logical_operator: and metadata_filtering_mode: manual multiple_retrieval_config: reranking_enable: false reranking_mode: weighted_score reranking_model: model: '' provider: '' score_threshold: null top_k: 10 weights: keyword_setting: keyword_weight: 0 vector_setting: embedding_model_name: text-embedding-3-large embedding_provider_name: langgenius/azure_openai/azure_openai vector_weight: 1 weight_type: customized query_variable_selector: - '17487852863990' - item retrieval_mode: multiple selected: false title: Knowledge Retrieval 5 type: knowledge-retrieval height: 90 id: '1748785286399017487852864000' parentId: '17487852863990' position: x: 213.39028285937735 y: 65 positionAbsolute: x: 669.6446491620802 y: 808.564737158242 selected: false sourcePosition: right targetPosition: left type: custom width: 242 zIndex: 1001 - data: desc: '' isInIteration: true selected: false title: '' type: iteration-start draggable: false height: 48 id: 1748785286399start parentId: '17487852863990' position: x: 24 y: 68 positionAbsolute: x: 480.2543663027028 y: 811.564737158242 selectable: false sourcePosition: right targetPosition: left type: custom-iteration-start width: 44 zIndex: 1001 - data: cases: - case_id: 'true' conditions: - comparison_operator: not empty id: 1f00de1d-b3c7-4f97-8409-165cb5d04016 value: '' varType: array[string] variable_selector: - '17487850509710' - issue_keys id: 'true' logical_operator: and - case_id: f2be164b-f0ce-4db6-9a7b-69786c02a988 conditions: - comparison_operator: empty id: 94834777-b870-481f-9bce-d5977fd70fe7 value: '' varType: array[string] variable_selector: - '17487850509710' - issue_keys id: f2be164b-f0ce-4db6-9a7b-69786c02a988 logical_operator: and desc: '' selected: false title: IF/ELSE type: if-else height: 204 id: '1748785479296' position: x: 89.0778811944802 y: 754.9250298061089 positionAbsolute: x: 89.0778811944802 y: 754.9250298061089 selected: false sourcePosition: right targetPosition: left type: custom width: 242 zIndex: 0 - data: context: enabled: true variable_selector: - sys - query desc: '' model: completion_params: temperature: 0.7 mode: chat name: gpt-35-turbo-16k provider: langgenius/azure_openai/azure_openai prompt_template: - id: f1e90d8d-abc0-478c-8604-f2805bc0c2f1 role: system text: "You are an experienced Test Case Generator.\n\n Your\ \ task is to create Gherkin-style test cases based on a user-reported\ \ requirement or issue described in natural language.\n The\ \ requirement may reference features, behaviors, bugs, or changes not\ \ currently in the knowledge base. You must rely solely on the user’s\ \ input and any additional context in the flow.\n\n \n\ \ {{#context#}}\n \n\n \ \ User requirement:\n {{#sys.query#}}\n\n \ \ ---\n\n \U0001F50D Once you have the requirement details,\ \ do the following:\n 1. Analyze the user input and extract\ \ the intent, preconditions, actions, and expected outcomes.\n \ \ 2. Generate one or more test cases using the Gherkin format.\n\ \n ---\n\n \U0001F4CA Test Case Format:\n\n\ \ ```gherkin\n Feature: [Concise feature name]\n\ \ Scenario: [High-level description of the scenario]\n\ \ Given [Initial precondition]\n And\ \ [Optional second precondition]\n When [Action performed\ \ by the user or system]\n And [Optional secondary action]\n\ \ Then [Expected outcome or behavior]\n \ \ And [Optional second expected outcome]\n ```\n\n \ \ ---\n\n ✅ Guidelines:\n - Use clear,\ \ concise, domain-appropriate language.\n - Focus on realistic\ \ testable behavior.\n - Avoid filler steps — each line should\ \ add functional value.\n - If multiple scenarios are implied,\ \ provide each in its own clearly labeled block.\n\n \U0001F4C6\ \ Present each test case inside a styled markdown code block, like this:\n\ \n ```gherkin\n Feature: Login security\n\n\ \ Scenario: User enters incorrect password\n \ \ Given the user is on the login page\n When the\ \ user enters a valid username\n And an incorrect password\n\ \ Then the user should see an error message\n \ \ And should not be logged in\n ```\n\n \ \ Do not include any explanation or commentary — only output the\ \ test case(s) in clean, Gherkin-formatted code blocks.\n" selected: false title: LLM 5 type: llm variables: [] vision: enabled: false height: 88 id: '1748785532843' position: x: 456.2543663027028 y: 986.9215676569963 positionAbsolute: x: 456.2543663027028 y: 986.9215676569963 selected: false sourcePosition: right targetPosition: left type: custom width: 242 zIndex: 0 - data: context: enabled: true variable_selector: - '1762728060345' - flat_context desc: '' model: completion_params: temperature: 0.7 mode: chat name: gpt-35-turbo-16k provider: langgenius/azure_openai/azure_openai prompt_template: - id: 94d687e6-c23d-43e3-977c-65cb9c4a8cc1 role: system text: "\nYou are an experienced Test Case Generator. Your task is to create\ \ test cases related to issues in context in Gherkin format based on\ \ the following user-reported issue (obtained from the knoledge retrieval\ \ node) and flow context:\n\n{{#context#}}\n\n{{#sys.query#}}\n\ Once you have the issue details, analyze the information, paying close\ \ attention to the title, description, and any steps to reproduce. \n\ Based on this analysis, generate a test case in Gherkin format using the\ \ following structure: \nFeature: [Concise title summarizing the feature\ \ being tested, derived from the Jira issue title] \n Scenario: [Specific\ \ scenario derived from the Jira issue details] \n - Given [Precondition\ \ 1, based on the issue context] \n - And [Precondition 2, if any] \n\ \ - When [Action 1 taken by the user, based on steps to reproduce] \n\ \ - And [Action 2 taken by the user, if any] \n - Then [Expected outcome\ \ based on the problem description and intended functionality] \n - And\ \ [Further expected outcome, if any] \n\nPlease ensure the Gherkin steps\ \ are clear, concise, and directly relate to the Jira issue.\nProvide\ \ the test cases in a visual way easy to identify, leave clear which is\ \ each scenario (with colours and inside a \"box\" as it was code)\n\n\ \n\nBatch completion status: {{#1762728060345.batch_status#}}\nIf this\ \ status is partial or empty, state it clearly before presenting any successful\ \ results." selected: false title: LLM 6 type: llm variables: [] vision: enabled: false height: 88 id: '1748785557565' position: x: 1434.2543663027027 y: 743.564737158242 positionAbsolute: x: 1434.2543663027027 y: 743.564737158242 selected: true sourcePosition: right targetPosition: left type: custom width: 242 zIndex: 0 - data: answer: '{{#1748785557565.text#}}' desc: '' selected: false title: Answer 6 type: answer variables: [] height: 103 id: '1748785836978' position: x: 1738.2543663027027 y: 743.564737158242 positionAbsolute: x: 1738.2543663027027 y: 743.564737158242 sourcePosition: right targetPosition: left type: custom width: 242 zIndex: 0 - data: answer: '{{#1748785532843.text#}}' desc: '' selected: false title: Answer 7 type: answer variables: [] height: 103 id: '1748786521492' position: x: 760.2543663027028 y: 986.9215676569963 positionAbsolute: x: 760.2543663027028 y: 986.9215676569963 sourcePosition: right targetPosition: left type: custom width: 242 zIndex: 0 - data: context: enabled: true variable_selector: - sys - query desc: '' model: completion_params: temperature: 0.7 mode: chat name: gpt-35-turbo-16k provider: langgenius/azure_openai/azure_openai prompt_template: - id: 3c1b3adc-6c0e-4224-a8e1-18c5f6c6fb6d role: system text: "Since this query {{#sys.query#}} is not related to the chatbot context\ \ it should return one of these answers:\ndefault_non_testing_responses:\n\ \ - \"I'm here to help with software testing topics. Could you rephrase\ \ your question to focus on testing or technical documentation?\"\n -\ \ \"This assistant is specialized in analyzing test cases, requirements,\ \ and project documentation. That question might be better suited for\ \ a general assistant.\"\n - \"I couldn’t find testing-relevant content\ \ in your question. If you’d like help with test generation or issue analysis,\ \ just let me know!\"\n - \"Hmm, I’m not sure how to assist with that.\ \ I specialize in software quality, requirements, and test design. Want\ \ to dig into a specific feature or issue?\"\n - \"This assistant works\ \ best with technical context. Try asking about a feature, bug, or test\ \ scenario you'd like to explore.\"\n" selected: false title: LLM 7 type: llm variables: [] vision: enabled: false height: 88 id: '1748786653855' position: x: -284.44013680327953 y: 1011.2915512785755 positionAbsolute: x: -284.44013680327953 y: 1011.2915512785755 selected: false sourcePosition: right targetPosition: left type: custom width: 242 zIndex: 0 - data: code: "import re\n\n\ndef main(inputs: list, expected_keys: list) -> dict:\n\ \ flat_context = []\n for item in inputs:\n if isinstance(item,\ \ list):\n flat_context.extend(value for value in item if isinstance(value,\ \ dict))\n elif isinstance(item, dict):\n flat_context.append(item)\n\ \n expected = [str(key).upper() for key in expected_keys if key]\n \ \ retrieved_keys = set(re.findall(\n r\"(? dict:\n\ \ flat_context = []\n for item in inputs:\n if isinstance(item,\ \ list):\n flat_context.extend(value for value in item if isinstance(value,\ \ dict))\n elif isinstance(item, dict):\n flat_context.append(item)\n\ \n expected = [str(key).upper() for key in expected_keys if key]\n \ \ retrieved_keys = set(re.findall(\n r\"(?