feat(i18n): Enhance internationalization support for tool specifications and notifications
- Added TitleKey and DescriptionKey fields to ToolSpec for dynamic localization. - Updated tool specifications to use i18nx for titles, descriptions, and appendices. - Refactored notification messages to utilize i18nx for localization in conversation and ticket assignment events. - Improved localization in the debug dialog component for skill definitions. - Added new translation keys in English and Chinese for ticket and conversation assignment notifications. - Ensured that fallback messages are properly localized based on user locale.
This commit is contained in:
mlogclub
@@ -389,3 +389,117 @@ error.agentTeamSchedule.startTimeInvalid: "Invalid start time format."
|
||||
error.agentTeamSchedule.startTimeInvalidWithFormat: "Invalid start time format. Use HH:mm or HH:mm:ss."
|
||||
error.agentTeamSchedule.endTimeInvalid: "Invalid end time format."
|
||||
error.agentTeamSchedule.endTimeInvalidWithFormat: "Invalid end time format. Use HH:mm or HH:mm:ss."
|
||||
tool.builtin.toolSearch.title: "Search and Run Dynamic Tools"
|
||||
tool.builtin.toolSearch.description: "Searches the MCP tools currently available to the agent and runs the selected tool after its toolCode is confirmed. Best for long-tail tools; it should not replace fixed built-in workflow tools."
|
||||
tool.builtin.toolSearch.appendix: |-
|
||||
When you need long-tail MCP capabilities, prefer the tool_search tool and follow these rules:
|
||||
1. Call tool_search first to find the dynamic tool you need, then continue with the selected real tool.
|
||||
2. Do not assume all long-tail tools are visible at the start; only tools selected by tool_search are exposed to subsequent model calls.
|
||||
3. If a fixed built-in tool can already complete the task, prefer the fixed tool and do not overuse tool_search.
|
||||
tool.builtin.skill.title: "Load Skill Instructions"
|
||||
tool.builtin.skill.description: "Loads specialized skill instructions for the current agent when extra task-specific guidance is needed."
|
||||
tool.graph.triageServiceRequest.title: "Route Service Request"
|
||||
tool.graph.triageServiceRequest.description: "Analyzes the current conversation to decide whether to keep answering, prepare a ticket draft, or hand off to a human, including a ticket draft when ticket creation is appropriate."
|
||||
tool.graph.triageServiceRequest.appendix: |-
|
||||
When you need to decide between continuing the answer, creating a ticket, or handing off to a human, call triage_service_request first and follow these rules:
|
||||
1. The tool returns recommendedAction and includes ticketDraft when ticket creation is needed.
|
||||
2. If recommendedAction=continue_answering, continue clarifying or answering instead of escalating directly.
|
||||
3. If recommendedAction=prepare_ticket, use ticketDraft or collect missing fields before calling create_ticket_with_confirmation.
|
||||
4. If recommendedAction=handoff_to_human, confirm the reason is sufficient before calling handoff_to_human.
|
||||
5. When the escalation path is unclear, use this tool instead of making a complex routing decision from the main prompt alone.
|
||||
tool.graph.analyzeConversation.title: "Analyze Conversation Risk and Summary"
|
||||
tool.graph.analyzeConversation.description: "Summarizes the current conversation, identifies risk signals, and recommends whether to keep answering, create a ticket, or hand off to a human."
|
||||
tool.graph.analyzeConversation.appendix: |-
|
||||
When the conversation may involve escalation, refunds, compensation, clear negative sentiment, ticket creation, or human handoff, call analyze_conversation first and follow these rules:
|
||||
1. This tool returns a structured summary, risk signals, and next-step recommendation. It does not create tickets or hand off to a human.
|
||||
2. If the tool recommends handoff_to_human, confirm the handoff conditions before calling handoff_to_human.
|
||||
3. If the tool recommends prepare_ticket, call prepare_ticket_draft or collect more information before creating a ticket.
|
||||
4. If the tool recommends continue_answering, continue clarifying and answering instead of escalating too early.
|
||||
tool.graph.prepareTicketDraft.title: "Prepare Ticket Draft"
|
||||
tool.graph.prepareTicketDraft.description: "Turns the current conversation and collected details into a ticket draft with a suggested title, description, missing fields, and follow-up questions."
|
||||
tool.graph.prepareTicketDraft.appendix: |-
|
||||
When the user has asked to create a ticket, file a complaint, report an issue, or request after-sales handling, but the title or description is still unclear, call prepare_ticket_draft first and follow these rules:
|
||||
1. This tool prepares a ticket draft and returns a suggested title, suggested description, missing fields, and follow-up questions.
|
||||
2. If ready=false, ask follow-up questions based on missingFields and followUpQuestions instead of creating a ticket directly.
|
||||
3. If ready=true, use the result to consider calling create_ticket_with_confirmation.
|
||||
4. This tool only prepares a draft. It does not create a ticket.
|
||||
tool.graph.createTicketConfirm.title: "Create Ticket With Confirmation"
|
||||
tool.graph.createTicketConfirm.description: "Guides ticket creation with parameter preparation, customer confirmation, actual ticket creation, and final result delivery."
|
||||
tool.graph.createTicketConfirm.appendix: |-
|
||||
You can call create_ticket_with_confirmation after enough information has been collected, but follow these rules:
|
||||
1. Only consider this tool when the user explicitly wants to submit a ticket, complaint, issue report, or after-sales request.
|
||||
2. Before calling it, prepare a clear ticket title and issue description. If the information is still scattered, call prepare_ticket_draft or ask follow-up questions first.
|
||||
3. Once you are ready to create a ticket, you must call create_ticket_with_confirmation. Do not simply claim in text that the ticket has been created.
|
||||
4. This Graph Tool asks the user for confirmation first. The ticket is created only after the user confirms; if the user cancels, the flow ends.
|
||||
5. If the user is only asking questions, complaining generally, or expressing dissatisfaction without explicitly requesting a ticket, continue clarifying instead of proactively creating one.
|
||||
tool.graph.handoffConversation.title: "Handoff to Human With Confirmation"
|
||||
tool.graph.handoffConversation.description: "Guides human handoff with reason preparation, customer confirmation, actual transfer, and final result delivery."
|
||||
tool.graph.handoffConversation.appendix: |-
|
||||
You can call handoff_to_human after confirming that human help is needed, but follow these rules:
|
||||
1. Only call this tool when the user explicitly asks for a human agent or you have determined that the issue must be handled by a human.
|
||||
2. Before calling it, summarize the handoff reason clearly. If the reason is vague, ask a follow-up question first.
|
||||
3. Once you decide to hand off, you must call handoff_to_human. Do not simply say in text that you have connected the user to a human.
|
||||
4. This Graph Tool asks the user for confirmation first. The handoff happens only after the user confirms; if the user cancels, the flow ends.
|
||||
5. If the issue can still be solved in the current conversation, continue helping instead of escalating too early.
|
||||
6. If the tool returns terminal=true and shouldRetry=false, the handoff flow has ended. Do not call it repeatedly.
|
||||
conversation.handoff.waiting: "We are connecting you to a human support agent. Please wait."
|
||||
conversation.handoff.offHours: "Human support is currently outside service hours. You can keep describing the issue and I will do my best to help. You can also request a human agent again when service hours resume."
|
||||
notification.ticketAssigned.title: "Ticket assigned"
|
||||
notification.ticketAssigned.line: "Ticket %s has been assigned to you."
|
||||
notification.ticketAssigned.reason: "Assignment reason: %s"
|
||||
notification.conversationTransferred.title: "Conversation transferred"
|
||||
notification.conversationAutoAssigned.title: "Conversation auto-assigned"
|
||||
notification.conversationAssigned.title: "Conversation assigned"
|
||||
notification.conversationAssigned.line: "Conversation #%s has been assigned to you."
|
||||
notification.conversationAssigned.reason: "Assignment reason: %s"
|
||||
notification.conversationTransferred.reason: "Transfer reason: %s"
|
||||
notification.ticketAssigned.wxwork.no: "Ticket no: %s"
|
||||
notification.ticketAssigned.wxwork.title: "Title: %s"
|
||||
notification.ticketAssigned.wxwork.status: "Status: %s"
|
||||
notification.assignee: "Assignee: %s"
|
||||
notification.conversationAssigned.wxwork.id: "Conversation ID: #%d"
|
||||
notification.conversationAssigned.wxwork.summary: "Summary: %s"
|
||||
notification.conversationAssigned.wxwork.channel: "Channel: %s"
|
||||
notification.conversationAssigned.wxwork.status: "Status: %s"
|
||||
notification.time: "Time: %s"
|
||||
graph.confirmOrCancel: "Please reply with \"Confirm\" or \"Cancel\"."
|
||||
graph.needExplicitConfirmation: "I need your explicit confirmation. Please reply with \"Confirm\" or \"Cancel\"."
|
||||
graph.confirmationExpired: "This confirmation has expired. Please start again."
|
||||
graph.cancelCreateTicket: "Ticket creation has been cancelled."
|
||||
graph.cancelHandoff: "Human handoff has been cancelled."
|
||||
graph.ticketCreated: "Ticket created. Ticket no: %s. Title: %s."
|
||||
graph.createTicketConfirmPrompt: |-
|
||||
I am ready to create a ticket for you.
|
||||
Title: %s
|
||||
Description: %s
|
||||
Please reply with "Confirm" or "Cancel".
|
||||
graph.defaultHandoffReason: "The user needs human support."
|
||||
graph.handoffConfirmPrompt: |-
|
||||
I am ready to connect you to a human support agent.
|
||||
Reason: %s
|
||||
Please reply with "Confirm" or "Cancel".
|
||||
conversation.interrupt.defaultPrompt: "Please provide more information and try again."
|
||||
ticket.defaultConversationTitle: "Conversation ticket"
|
||||
tool.graph.createTicketConfirm.info: "Graph Tool. Handles ticket parameter preparation, user confirmation, actual ticket creation, and result return. Use only when the user explicitly asks to create a ticket and the title and description are clear."
|
||||
tool.graph.createTicketConfirm.param.title: "Ticket title. Concisely summarizes the issue."
|
||||
tool.graph.createTicketConfirm.param.description: "Ticket description. Clearly captures the user's issue, symptoms, and request."
|
||||
tool.graph.handoffConversation.info: "Graph Tool. Handles handoff reason preparation, user confirmation, actual human handoff, and result return. Use only when the user explicitly asks for a human agent or you have confirmed that human handling is required. Do not repeat the call when the result has terminal=true and shouldRetry=false."
|
||||
tool.graph.handoffConversation.param.reason: "Handoff reason. Briefly explain why a human is needed, such as explicit user request, manual verification, or after-sales handling."
|
||||
tool.graph.triageServiceRequest.info: "Graph Tool. Analyzes the current conversation to decide whether to continue answering, prepare a ticket draft, or hand off to a human. When ticket creation is recommended, it returns a structured ticket draft suggestion."
|
||||
tool.graph.triageServiceRequest.param.goal: "Analysis goal, such as whether to escalate, create a ticket, or hand off to a human."
|
||||
tool.graph.triageServiceRequest.param.observedIssue: "Main issue or dispute observed in the conversation."
|
||||
tool.graph.triageServiceRequest.param.needTicket: "Whether to focus on evaluating ticket creation."
|
||||
tool.graph.triageServiceRequest.param.needHumanHandoff: "Whether to focus on evaluating human handoff."
|
||||
tool.graph.triageServiceRequest.param.additionalContext: "Additional context, such as risk signals or constraints already identified."
|
||||
tool.graph.analyzeConversation.info: "Graph Tool. Summarizes the current conversation, identifies complaint/payment/sentiment risk signals, and recommends whether to continue answering, create a ticket, or hand off to a human."
|
||||
tool.graph.analyzeConversation.param.goal: "Analysis goal, such as whether to hand off to a human, create a ticket, or perform risk review."
|
||||
tool.graph.analyzeConversation.param.observedIssue: "Main issue or request observed in the conversation."
|
||||
tool.graph.analyzeConversation.param.needQualityCheck: "Whether to focus on risk or quality review."
|
||||
tool.graph.analyzeConversation.param.additionalContext: "Additional context, such as disputes, complaint points, or business constraints already identified."
|
||||
tool.graph.prepareTicketDraft.info: "Graph Tool. Prepares a ticket draft from the current conversation and collected information. Use it before create_ticket_with_confirmation when the ticket content needs to be organized."
|
||||
tool.graph.prepareTicketDraft.param.title: "Prepared ticket title. Optional."
|
||||
tool.graph.prepareTicketDraft.param.description: "Prepared ticket description. Optional."
|
||||
tool.graph.prepareTicketDraft.param.issue: "The issue or error message the user is experiencing."
|
||||
tool.graph.prepareTicketDraft.param.impact: "Impact scope, such as unable to sign in, unable to place an order, or business interruption."
|
||||
tool.graph.prepareTicketDraft.param.expectedOutcome: "The user's expected outcome or request."
|
||||
tool.graph.prepareTicketDraft.param.currentAttempt: "Current attempted troubleshooting steps. Optional."
|
||||
|
||||
@@ -389,3 +389,117 @@ error.agentTeamSchedule.startTimeInvalid: "开始时间格式错误"
|
||||
error.agentTeamSchedule.startTimeInvalidWithFormat: "开始时间格式错误，请使用 HH:mm 或 HH:mm:ss"
|
||||
error.agentTeamSchedule.endTimeInvalid: "结束时间格式错误"
|
||||
error.agentTeamSchedule.endTimeInvalidWithFormat: "结束时间格式错误，请使用 HH:mm 或 HH:mm:ss"
|
||||
tool.builtin.toolSearch.title: "搜索并调用动态工具"
|
||||
tool.builtin.toolSearch.description: "用于搜索当前允许使用的 MCP 工具，并在确认目标 toolCode 后动态调用该工具。适合处理长尾工具，不应替代固定内置流程工具。"
|
||||
tool.builtin.toolSearch.appendix: |-
|
||||
当你需要使用长尾 MCP 能力时，优先使用 tool_search 工具，并遵守以下规则：
|
||||
1. 先调用 tool_search 搜索需要的动态工具，再继续使用已选中的真实工具。
|
||||
2. 不要假设所有长尾工具一开始就可见；只有被 tool_search 选中的工具，后续模型调用才会暴露出来。
|
||||
3. 如果当前已有固定内置工具可以完成任务，优先使用固定工具，不要滥用 tool_search。
|
||||
tool.builtin.skill.title: "加载专项技能说明"
|
||||
tool.builtin.skill.description: "用于按需加载当前 Agent 可用的专项技能说明文档，适合在需要专项处理规则时再注入上下文。"
|
||||
tool.graph.triageServiceRequest.title: "服务请求分流"
|
||||
tool.graph.triageServiceRequest.description: "Graph Tool。分析当前会话，判断应继续回答、准备工单草稿还是转人工；当适合创建工单时，也会准备工单草稿。"
|
||||
tool.graph.triageServiceRequest.appendix: |-
|
||||
当你需要在继续回答、创建工单、转人工之间做判断时，优先调用 triage_service_request，并遵守以下规则：
|
||||
1. 工具会返回 recommendedAction，并在需要创建工单时包含 ticketDraft。
|
||||
2. 如果 recommendedAction=continue_answering，继续澄清或回答，不要直接升级。
|
||||
3. 如果 recommendedAction=prepare_ticket，使用 ticketDraft 或补充缺失字段后，再考虑调用 create_ticket_with_confirmation。
|
||||
4. 如果 recommendedAction=handoff_to_human，先确认转人工原因足够明确，再调用 handoff_to_human。
|
||||
5. 当升级路径不明确时，使用此工具，而不是只依赖主提示词做复杂路由判断。
|
||||
tool.graph.analyzeConversation.title: "分析会话风险和摘要"
|
||||
tool.graph.analyzeConversation.description: "Graph Tool。总结当前会话，识别风险信号，并建议继续回答、创建工单或转人工。"
|
||||
tool.graph.analyzeConversation.appendix: |-
|
||||
当会话可能涉及升级、退款、赔偿、明显负面情绪、创建工单或转人工时，优先调用 analyze_conversation，并遵守以下规则：
|
||||
1. 此工具返回结构化摘要、风险信号和下一步建议；它不会创建工单或执行转人工。
|
||||
2. 如果工具建议 handoff_to_human，先确认转人工条件，再调用 handoff_to_human。
|
||||
3. 如果工具建议 prepare_ticket，调用 prepare_ticket_draft 或继续收集信息后再创建工单。
|
||||
4. 如果工具建议 continue_answering，继续澄清和回答，不要过早升级。
|
||||
tool.graph.prepareTicketDraft.title: "准备工单草稿"
|
||||
tool.graph.prepareTicketDraft.description: "Graph Tool。基于当前会话和已收集信息准备工单草稿，包括建议标题、描述、缺失字段和追问问题。"
|
||||
tool.graph.prepareTicketDraft.appendix: |-
|
||||
当用户要求创建工单、投诉、报障或售后处理，但标题或描述仍不清晰时，优先调用 prepare_ticket_draft，并遵守以下规则：
|
||||
1. 此工具会准备工单草稿，并返回建议标题、建议描述、缺失字段和追问问题。
|
||||
2. 如果 ready=false，根据 missingFields 和 followUpQuestions 继续追问，不要直接创建工单。
|
||||
3. 如果 ready=true，使用结果考虑调用 create_ticket_with_confirmation。
|
||||
4. 此工具只准备草稿，不会创建工单。
|
||||
tool.graph.createTicketConfirm.title: "创建工单确认流程"
|
||||
tool.graph.createTicketConfirm.description: "Graph Tool。处理工单参数准备、用户确认、实际创建工单和结果返回。"
|
||||
tool.graph.createTicketConfirm.appendix: |-
|
||||
收集到足够信息后，可以调用 create_ticket_with_confirmation，但必须遵守以下规则：
|
||||
1. 只有当用户明确想提交工单、投诉、问题报告或售后请求时，才考虑使用此工具。
|
||||
2. 调用前准备清晰的工单标题和问题描述；如果信息仍然分散，先调用 prepare_ticket_draft 或继续追问。
|
||||
3. 一旦准备创建工单，必须调用 create_ticket_with_confirmation，不要只用文本声称工单已创建。
|
||||
4. 此 Graph Tool 会先请求用户确认；只有用户确认后才会创建工单，用户取消则流程结束。
|
||||
5. 如果用户只是提问、泛泛抱怨或表达不满但没有明确要求创建工单，应继续澄清，而不是主动创建。
|
||||
tool.graph.handoffConversation.title: "转人工确认流程"
|
||||
tool.graph.handoffConversation.description: "Graph Tool。处理转人工原因准备、用户确认、实际转人工和结果返回。"
|
||||
tool.graph.handoffConversation.appendix: |-
|
||||
确认需要人工帮助后，可以调用 handoff_to_human，但必须遵守以下规则：
|
||||
1. 只有当用户明确要求人工客服，或你判断该问题必须由人工处理时才调用此工具。
|
||||
2. 调用前清晰总结转人工原因；如果原因模糊，先追问。
|
||||
3. 一旦决定转人工，必须调用 handoff_to_human，不要只用文本声称已接入人工。
|
||||
4. 此 Graph Tool 会先请求用户确认；只有用户确认后才会转人工，用户取消则流程结束。
|
||||
5. 如果问题仍可在当前会话解决，继续帮助用户，不要过早升级。
|
||||
6. 如果工具返回 terminal=true 且 shouldRetry=false，说明转人工流程已结束，不要重复调用。
|
||||
conversation.handoff.waiting: "正在为你接入人工客服，请稍候。"
|
||||
conversation.handoff.offHours: "当前不在人工客服服务时间内。你可以继续描述问题，我会尽力协助；也可以在服务时间恢复后再次申请人工客服。"
|
||||
notification.ticketAssigned.title: "工单指派提醒"
|
||||
notification.ticketAssigned.line: "工单 %s 已指派给你"
|
||||
notification.ticketAssigned.reason: "指派原因: %s"
|
||||
notification.conversationTransferred.title: "会话转接提醒"
|
||||
notification.conversationAutoAssigned.title: "会话自动分配提醒"
|
||||
notification.conversationAssigned.title: "会话分配提醒"
|
||||
notification.conversationAssigned.line: "会话 #%s 已分配给你"
|
||||
notification.conversationAssigned.reason: "分配原因: %s"
|
||||
notification.conversationTransferred.reason: "转接原因: %s"
|
||||
notification.ticketAssigned.wxwork.no: "工单号: %s"
|
||||
notification.ticketAssigned.wxwork.title: "工单标题: %s"
|
||||
notification.ticketAssigned.wxwork.status: "当前状态: %s"
|
||||
notification.assignee: "处理人: %s"
|
||||
notification.conversationAssigned.wxwork.id: "会话ID: #%d"
|
||||
notification.conversationAssigned.wxwork.summary: "摘要: %s"
|
||||
notification.conversationAssigned.wxwork.channel: "渠道: %s"
|
||||
notification.conversationAssigned.wxwork.status: "状态: %s"
|
||||
notification.time: "时间: %s"
|
||||
graph.confirmOrCancel: "请回复“确认”或“取消”。"
|
||||
graph.needExplicitConfirmation: "需要你明确确认。请回复“确认”或“取消”。"
|
||||
graph.confirmationExpired: "本次确认已过期，请重新发起。"
|
||||
graph.cancelCreateTicket: "已取消本次工单创建。"
|
||||
graph.cancelHandoff: "已取消本次转人工。"
|
||||
graph.ticketCreated: "工单已创建。工单号：%s。标题：%s。"
|
||||
graph.createTicketConfirmPrompt: |-
|
||||
我已准备好为你创建工单。
|
||||
标题：%s
|
||||
描述：%s
|
||||
请回复“确认”或“取消”。
|
||||
graph.defaultHandoffReason: "用户需要人工客服支持。"
|
||||
graph.handoffConfirmPrompt: |-
|
||||
我已准备好为你接入人工客服。
|
||||
原因：%s
|
||||
请回复“确认”或“取消”。
|
||||
conversation.interrupt.defaultPrompt: "请补充更多信息后再试。"
|
||||
ticket.defaultConversationTitle: "会话工单"
|
||||
tool.graph.createTicketConfirm.info: "Graph Tool。处理工单参数准备、用户确认、实际创建工单和结果返回。仅当用户明确要求创建工单且标题、描述清晰时使用。"
|
||||
tool.graph.createTicketConfirm.param.title: "工单标题。简洁概括问题。"
|
||||
tool.graph.createTicketConfirm.param.description: "工单描述。清晰记录用户的问题、现象和诉求。"
|
||||
tool.graph.handoffConversation.info: "Graph Tool。处理转人工原因准备、用户确认、实际转人工和结果返回。仅当用户明确要求人工客服，或你确认需要人工处理时使用。当结果 terminal=true 且 shouldRetry=false 时不要重复调用。"
|
||||
tool.graph.handoffConversation.param.reason: "转人工原因。简要说明为什么需要人工，例如用户明确要求、需要人工核验或售后处理。"
|
||||
tool.graph.triageServiceRequest.info: "Graph Tool。分析当前会话，判断应继续回答、准备工单草稿还是转人工。建议创建工单时，会返回结构化工单草稿建议。"
|
||||
tool.graph.triageServiceRequest.param.goal: "分析目标，例如是否升级、创建工单或转人工。"
|
||||
tool.graph.triageServiceRequest.param.observedIssue: "当前会话中观察到的主要问题或争议。"
|
||||
tool.graph.triageServiceRequest.param.needTicket: "是否重点评估创建工单。"
|
||||
tool.graph.triageServiceRequest.param.needHumanHandoff: "是否重点评估转人工。"
|
||||
tool.graph.triageServiceRequest.param.additionalContext: "补充上下文，例如已经识别的风险信号或约束。"
|
||||
tool.graph.analyzeConversation.info: "Graph Tool。总结当前会话，识别投诉、支付、情绪等风险信号，并建议继续回答、创建工单或转人工。"
|
||||
tool.graph.analyzeConversation.param.goal: "分析目标，例如是否转人工、创建工单或执行风险复核。"
|
||||
tool.graph.analyzeConversation.param.observedIssue: "当前会话中观察到的主要问题或请求。"
|
||||
tool.graph.analyzeConversation.param.needQualityCheck: "是否重点进行风险或质量复核。"
|
||||
tool.graph.analyzeConversation.param.additionalContext: "补充上下文，例如争议点、投诉点或已识别的业务约束。"
|
||||
tool.graph.prepareTicketDraft.info: "Graph Tool。根据当前会话和已收集信息准备工单草稿。当工单内容需要整理时，应在 create_ticket_with_confirmation 之前使用。"
|
||||
tool.graph.prepareTicketDraft.param.title: "准备好的工单标题，可选。"
|
||||
tool.graph.prepareTicketDraft.param.description: "准备好的工单描述，可选。"
|
||||
tool.graph.prepareTicketDraft.param.issue: "用户正在遇到的问题或报错信息。"
|
||||
tool.graph.prepareTicketDraft.param.impact: "影响范围，例如无法登录、无法下单或业务中断。"
|
||||
tool.graph.prepareTicketDraft.param.expectedOutcome: "用户期望的处理结果或诉求。"
|
||||
tool.graph.prepareTicketDraft.param.currentAttempt: "当前已尝试过的处理步骤，可选。"
|
||||
|
||||
+114
-162
@@ -8,17 +8,20 @@ import (
|
||||
)
|
||||
|
||||
type ToolSpec struct {
|
||||
Code string
|
||||
ServerCode string
|
||||
Name string
|
||||
Title string
|
||||
Description string
|
||||
SourceType enums.ToolSourceType
|
||||
AutoInjected bool
|
||||
DirectAccess bool
|
||||
RuntimeStatic bool
|
||||
Aliases []string
|
||||
Appendix string
|
||||
Code string
|
||||
ServerCode string
|
||||
Name string
|
||||
Title string
|
||||
TitleKey string
|
||||
Description string
|
||||
DescriptionKey string
|
||||
SourceType enums.ToolSourceType
|
||||
AutoInjected bool
|
||||
DirectAccess bool
|
||||
RuntimeStatic bool
|
||||
Aliases []string
|
||||
Appendix string
|
||||
AppendixKey string
|
||||
}
|
||||
|
||||
type ToolMetadata struct {
|
||||
@@ -30,116 +33,97 @@ type ToolMetadata struct {
|
||||
|
||||
var (
|
||||
BuiltinToolSearch = ToolSpec{
|
||||
Code: "builtin/tool_search",
|
||||
ServerCode: "builtin",
|
||||
Name: "tool_search",
|
||||
Title: "搜索并调用动态工具",
|
||||
Description: "用于搜索当前允许使用的 MCP 工具，并在确认目标 toolCode 后动态调用该工具。适合处理长尾工具，不应替代固定内置流程工具。",
|
||||
SourceType: enums.ToolSourceTypeBuiltin,
|
||||
AutoInjected: true,
|
||||
DirectAccess: true,
|
||||
Appendix: strings.TrimSpace(`
|
||||
当你需要使用长尾 MCP 能力时，优先使用 tool_search 工具，并遵守以下规则：
|
||||
1. 先调用 tool_search 搜索需要的动态工具，再继续使用已选中的真实工具。
|
||||
2. 不要假设所有长尾工具一开始就可见；只有被 tool_search 选中的工具，后续模型调用才会暴露出来。
|
||||
3. 如果当前已有固定内置工具可以完成任务，优先使用固定工具，不要滥用 tool_search。
|
||||
`),
|
||||
Code: "builtin/tool_search",
|
||||
ServerCode: "builtin",
|
||||
Name: "tool_search",
|
||||
Title: i18nx.Get("tool.builtin.toolSearch.title"),
|
||||
TitleKey: "tool.builtin.toolSearch.title",
|
||||
Description: i18nx.Get("tool.builtin.toolSearch.description"),
|
||||
DescriptionKey: "tool.builtin.toolSearch.description",
|
||||
SourceType: enums.ToolSourceTypeBuiltin,
|
||||
AutoInjected: true,
|
||||
DirectAccess: true,
|
||||
Appendix: i18nx.Get("tool.builtin.toolSearch.appendix"),
|
||||
AppendixKey: "tool.builtin.toolSearch.appendix",
|
||||
}
|
||||
BuiltinSkill = ToolSpec{
|
||||
Code: "builtin/skill",
|
||||
ServerCode: "builtin",
|
||||
Name: "skill",
|
||||
Title: "加载专项技能说明",
|
||||
Description: "用于按需加载当前 Agent 可用的专项技能说明文档，适合在需要专项处理规则时再注入上下文。",
|
||||
SourceType: enums.ToolSourceTypeBuiltin,
|
||||
AutoInjected: true,
|
||||
Code: "builtin/skill",
|
||||
ServerCode: "builtin",
|
||||
Name: "skill",
|
||||
Title: i18nx.Get("tool.builtin.skill.title"),
|
||||
TitleKey: "tool.builtin.skill.title",
|
||||
Description: i18nx.Get("tool.builtin.skill.description"),
|
||||
DescriptionKey: "tool.builtin.skill.description",
|
||||
SourceType: enums.ToolSourceTypeBuiltin,
|
||||
AutoInjected: true,
|
||||
}
|
||||
GraphTriageServiceRequest = ToolSpec{
|
||||
Code: "graph/triage_service_request",
|
||||
ServerCode: "graph",
|
||||
Name: "triage_service_request",
|
||||
Title: "Triage service request",
|
||||
Description: "Graph Tool. Analyzes the current conversation to decide whether to continue answering, prepare a ticket draft, or hand off to a human. It also prepares a ticket draft when ticket creation is recommended.",
|
||||
SourceType: enums.ToolSourceTypeGraph,
|
||||
RuntimeStatic: true,
|
||||
Appendix: strings.TrimSpace(`
|
||||
When you need to decide between continuing the answer, creating a ticket, or handing off to a human, call triage_service_request first and follow these rules:
|
||||
1. The tool returns recommendedAction and includes ticketDraft when ticket creation is needed.
|
||||
2. If recommendedAction=continue_answering, continue clarifying or answering instead of escalating directly.
|
||||
3. If recommendedAction=prepare_ticket, use ticketDraft or collect missing fields before calling create_ticket_with_confirmation.
|
||||
4. If recommendedAction=handoff_to_human, confirm the reason is sufficient before calling handoff_to_human.
|
||||
5. When the escalation path is unclear, use this tool instead of making a complex routing decision from the main prompt alone.
|
||||
`),
|
||||
Code: "graph/triage_service_request",
|
||||
ServerCode: "graph",
|
||||
Name: "triage_service_request",
|
||||
Title: i18nx.Get("tool.graph.triageServiceRequest.title"),
|
||||
TitleKey: "tool.graph.triageServiceRequest.title",
|
||||
Description: i18nx.Get("tool.graph.triageServiceRequest.description"),
|
||||
DescriptionKey: "tool.graph.triageServiceRequest.description",
|
||||
SourceType: enums.ToolSourceTypeGraph,
|
||||
RuntimeStatic: true,
|
||||
Appendix: i18nx.Get("tool.graph.triageServiceRequest.appendix"),
|
||||
AppendixKey: "tool.graph.triageServiceRequest.appendix",
|
||||
}
|
||||
GraphAnalyzeConversation = ToolSpec{
|
||||
Code: "graph/analyze_conversation",
|
||||
ServerCode: "graph",
|
||||
Name: "analyze_conversation",
|
||||
Title: "Analyze conversation risk and summary",
|
||||
Description: "Graph Tool. Summarizes the current conversation, identifies risk signals, and recommends whether to continue answering, create a ticket, or hand off to a human.",
|
||||
SourceType: enums.ToolSourceTypeGraph,
|
||||
RuntimeStatic: true,
|
||||
Appendix: strings.TrimSpace(`
|
||||
When the conversation may involve escalation, refunds, compensation, clear negative sentiment, ticket creation, or human handoff, call analyze_conversation first and follow these rules:
|
||||
1. This tool returns a structured summary, risk signals, and next-step recommendation. It does not create tickets or hand off to a human.
|
||||
2. If the tool recommends handoff_to_human, confirm the handoff conditions before calling handoff_to_human.
|
||||
3. If the tool recommends prepare_ticket, call prepare_ticket_draft or collect more information before creating a ticket.
|
||||
4. If the tool recommends continue_answering, continue clarifying and answering instead of escalating too early.
|
||||
`),
|
||||
Code: "graph/analyze_conversation",
|
||||
ServerCode: "graph",
|
||||
Name: "analyze_conversation",
|
||||
Title: i18nx.Get("tool.graph.analyzeConversation.title"),
|
||||
TitleKey: "tool.graph.analyzeConversation.title",
|
||||
Description: i18nx.Get("tool.graph.analyzeConversation.description"),
|
||||
DescriptionKey: "tool.graph.analyzeConversation.description",
|
||||
SourceType: enums.ToolSourceTypeGraph,
|
||||
RuntimeStatic: true,
|
||||
Appendix: i18nx.Get("tool.graph.analyzeConversation.appendix"),
|
||||
AppendixKey: "tool.graph.analyzeConversation.appendix",
|
||||
}
|
||||
GraphPrepareTicketDraft = ToolSpec{
|
||||
Code: "graph/prepare_ticket_draft",
|
||||
ServerCode: "graph",
|
||||
Name: "prepare_ticket_draft",
|
||||
Title: "Prepare ticket draft",
|
||||
Description: "Graph Tool. Prepares a ticket draft from the current conversation and collected information, including a suggested title, description, missing fields, and follow-up questions.",
|
||||
SourceType: enums.ToolSourceTypeGraph,
|
||||
RuntimeStatic: true,
|
||||
Appendix: strings.TrimSpace(`
|
||||
When the user has asked to create a ticket, file a complaint, report an issue, or request after-sales handling, but the title or description is still unclear, call prepare_ticket_draft first and follow these rules:
|
||||
1. This tool prepares a ticket draft and returns a suggested title, suggested description, missing fields, and follow-up questions.
|
||||
2. If ready=false, ask follow-up questions based on missingFields and followUpQuestions instead of creating a ticket directly.
|
||||
3. If ready=true, use the result to consider calling create_ticket_with_confirmation.
|
||||
4. This tool only prepares a draft. It does not create a ticket.
|
||||
`),
|
||||
Code: "graph/prepare_ticket_draft",
|
||||
ServerCode: "graph",
|
||||
Name: "prepare_ticket_draft",
|
||||
Title: i18nx.Get("tool.graph.prepareTicketDraft.title"),
|
||||
TitleKey: "tool.graph.prepareTicketDraft.title",
|
||||
Description: i18nx.Get("tool.graph.prepareTicketDraft.description"),
|
||||
DescriptionKey: "tool.graph.prepareTicketDraft.description",
|
||||
SourceType: enums.ToolSourceTypeGraph,
|
||||
RuntimeStatic: true,
|
||||
Appendix: i18nx.Get("tool.graph.prepareTicketDraft.appendix"),
|
||||
AppendixKey: "tool.graph.prepareTicketDraft.appendix",
|
||||
}
|
||||
GraphCreateTicketConfirm = ToolSpec{
|
||||
Code: "graph/create_ticket_with_confirmation",
|
||||
ServerCode: "graph",
|
||||
Name: "create_ticket_with_confirmation",
|
||||
Title: "Create ticket confirmation flow",
|
||||
Description: "Graph Tool. Handles ticket parameter preparation, user confirmation, actual ticket creation, and result return.",
|
||||
SourceType: enums.ToolSourceTypeGraph,
|
||||
DirectAccess: true,
|
||||
RuntimeStatic: true,
|
||||
Aliases: []string{"builtin/create_ticket_with_confirmation"},
|
||||
Appendix: strings.TrimSpace(`
|
||||
You can call create_ticket_with_confirmation after enough information has been collected, but follow these rules:
|
||||
1. Only consider this tool when the user explicitly wants to submit a ticket, complaint, issue report, or after-sales request.
|
||||
2. Before calling it, prepare a clear ticket title and issue description. If the information is still scattered, call prepare_ticket_draft or ask follow-up questions first.
|
||||
3. Once you are ready to create a ticket, you must call create_ticket_with_confirmation. Do not simply claim in text that the ticket has been created.
|
||||
4. This Graph Tool asks the user for confirmation first. The ticket is created only after the user confirms; if the user cancels, the flow ends.
|
||||
5. If the user is only asking questions, complaining generally, or expressing dissatisfaction without explicitly requesting a ticket, continue clarifying instead of proactively creating one.
|
||||
`),
|
||||
Code: "graph/create_ticket_with_confirmation",
|
||||
ServerCode: "graph",
|
||||
Name: "create_ticket_with_confirmation",
|
||||
Title: i18nx.Get("tool.graph.createTicketConfirm.title"),
|
||||
TitleKey: "tool.graph.createTicketConfirm.title",
|
||||
Description: i18nx.Get("tool.graph.createTicketConfirm.description"),
|
||||
DescriptionKey: "tool.graph.createTicketConfirm.description",
|
||||
SourceType: enums.ToolSourceTypeGraph,
|
||||
DirectAccess: true,
|
||||
RuntimeStatic: true,
|
||||
Aliases: []string{"builtin/create_ticket_with_confirmation"},
|
||||
Appendix: i18nx.Get("tool.graph.createTicketConfirm.appendix"),
|
||||
AppendixKey: "tool.graph.createTicketConfirm.appendix",
|
||||
}
|
||||
GraphHandoffConversation = ToolSpec{
|
||||
Code: "graph/handoff_to_human",
|
||||
ServerCode: "graph",
|
||||
Name: "handoff_to_human",
|
||||
Title: "Human handoff confirmation flow",
|
||||
Description: "Graph Tool. Handles handoff reason preparation, user confirmation, actual human handoff, and result return.",
|
||||
SourceType: enums.ToolSourceTypeGraph,
|
||||
DirectAccess: true,
|
||||
RuntimeStatic: true,
|
||||
Appendix: strings.TrimSpace(`
|
||||
You can call handoff_to_human after confirming that human help is needed, but follow these rules:
|
||||
1. Only call this tool when the user explicitly asks for a human agent or you have determined that the issue must be handled by a human.
|
||||
2. Before calling it, summarize the handoff reason clearly. If the reason is vague, ask a follow-up question first.
|
||||
3. Once you decide to hand off, you must call handoff_to_human. Do not simply say in text that you have connected the user to a human.
|
||||
4. This Graph Tool asks the user for confirmation first. The handoff happens only after the user confirms; if the user cancels, the flow ends.
|
||||
5. If the issue can still be solved in the current conversation, continue helping instead of escalating too early.
|
||||
6. If the tool returns terminal=true and shouldRetry=false, the handoff flow has ended. Do not call it repeatedly.
|
||||
`),
|
||||
Code: "graph/handoff_to_human",
|
||||
ServerCode: "graph",
|
||||
Name: "handoff_to_human",
|
||||
Title: i18nx.Get("tool.graph.handoffConversation.title"),
|
||||
TitleKey: "tool.graph.handoffConversation.title",
|
||||
Description: i18nx.Get("tool.graph.handoffConversation.description"),
|
||||
DescriptionKey: "tool.graph.handoffConversation.description",
|
||||
SourceType: enums.ToolSourceTypeGraph,
|
||||
DirectAccess: true,
|
||||
RuntimeStatic: true,
|
||||
Appendix: i18nx.Get("tool.graph.handoffConversation.appendix"),
|
||||
AppendixKey: "tool.graph.handoffConversation.appendix",
|
||||
}
|
||||
RegisteredToolSpecs = []ToolSpec{
|
||||
BuiltinToolSearch,
|
||||
@@ -224,11 +208,8 @@ func GetRegisteredToolTitleLocale(toolCode string, locale string) string {
|
||||
if !ok {
|
||||
return ""
|
||||
}
|
||||
if i18nx.NormalizeLocale(locale) != i18nx.LocaleEnUS {
|
||||
return spec.Title
|
||||
}
|
||||
if text := registeredToolEnglishTitle(spec.Code); text != "" {
|
||||
return text
|
||||
if strings.TrimSpace(spec.TitleKey) != "" {
|
||||
return i18nx.Getf(locale, spec.TitleKey)
|
||||
}
|
||||
return spec.Title
|
||||
}
|
||||
@@ -246,55 +227,21 @@ func GetRegisteredToolDescriptionLocale(toolCode string, locale string) string {
|
||||
if !ok {
|
||||
return ""
|
||||
}
|
||||
if i18nx.NormalizeLocale(locale) != i18nx.LocaleEnUS {
|
||||
return spec.Description
|
||||
}
|
||||
if text := registeredToolEnglishDescription(spec.Code); text != "" {
|
||||
return text
|
||||
if strings.TrimSpace(spec.DescriptionKey) != "" {
|
||||
return i18nx.Getf(locale, spec.DescriptionKey)
|
||||
}
|
||||
return spec.Description
|
||||
}
|
||||
|
||||
func registeredToolEnglishTitle(toolCode string) string {
|
||||
switch toolCode {
|
||||
case BuiltinToolSearch.Code:
|
||||
return "Search and Run Dynamic Tools"
|
||||
case BuiltinSkill.Code:
|
||||
return "Load Skill Instructions"
|
||||
case GraphTriageServiceRequest.Code:
|
||||
return "Route Service Request"
|
||||
case GraphAnalyzeConversation.Code:
|
||||
return "Analyze Conversation Risk and Summary"
|
||||
case GraphPrepareTicketDraft.Code:
|
||||
return "Prepare Ticket Draft"
|
||||
case GraphCreateTicketConfirm.Code:
|
||||
return "Create Ticket With Confirmation"
|
||||
case GraphHandoffConversation.Code:
|
||||
return "Handoff to Human With Confirmation"
|
||||
default:
|
||||
func GetRegisteredToolAppendixLocale(toolCode string, locale string) string {
|
||||
spec, ok := GetRegisteredToolSpec(toolCode)
|
||||
if !ok {
|
||||
return ""
|
||||
}
|
||||
}
|
||||
|
||||
func registeredToolEnglishDescription(toolCode string) string {
|
||||
switch toolCode {
|
||||
case BuiltinToolSearch.Code:
|
||||
return "Searches the MCP tools currently available to the agent and runs the selected tool after its toolCode is confirmed. Best for long-tail tools; it should not replace fixed built-in workflow tools."
|
||||
case BuiltinSkill.Code:
|
||||
return "Loads specialized skill instructions for the current agent when extra task-specific guidance is needed."
|
||||
case GraphTriageServiceRequest.Code:
|
||||
return "Analyzes the current conversation to decide whether to keep answering, prepare a ticket draft, or hand off to a human, including a ticket draft when ticket creation is appropriate."
|
||||
case GraphAnalyzeConversation.Code:
|
||||
return "Summarizes the current conversation, identifies risk signals, and recommends whether to keep answering, create a ticket, or hand off to a human."
|
||||
case GraphPrepareTicketDraft.Code:
|
||||
return "Turns the current conversation and collected details into a ticket draft with a suggested title, description, missing fields, and follow-up questions."
|
||||
case GraphCreateTicketConfirm.Code:
|
||||
return "Guides ticket creation with parameter preparation, customer confirmation, actual ticket creation, and final result delivery."
|
||||
case GraphHandoffConversation.Code:
|
||||
return "Guides human handoff with reason preparation, customer confirmation, actual transfer, and final result delivery."
|
||||
default:
|
||||
return ""
|
||||
if strings.TrimSpace(spec.AppendixKey) != "" {
|
||||
return i18nx.Getf(locale, spec.AppendixKey)
|
||||
}
|
||||
return spec.Appendix
|
||||
}
|
||||
|
||||
func GetRegisteredToolIdentity(toolCode string) (serverCode, toolName string, ok bool) {
|
||||
@@ -372,20 +319,25 @@ func BuildToolAppendices(hasDynamicMCPTools bool, toolCodes map[string]string) [
|
||||
}
|
||||
|
||||
func BuildToolAppendicesForCodes(hasDynamicMCPTools bool, toolCodes []string) []string {
|
||||
return BuildToolAppendicesForCodesLocale(hasDynamicMCPTools, toolCodes, i18nx.DefaultLocale)
|
||||
}
|
||||
|
||||
func BuildToolAppendicesForCodesLocale(hasDynamicMCPTools bool, toolCodes []string, locale string) []string {
|
||||
ret := make([]string, 0, len(toolCodes)+1)
|
||||
normalizedToolCodes := NormalizeToolCodes(toolCodes)
|
||||
if hasDynamicMCPTools && strings.TrimSpace(BuiltinToolSearch.Appendix) != "" {
|
||||
ret = append(ret, BuiltinToolSearch.Appendix)
|
||||
if hasDynamicMCPTools {
|
||||
if appendix := strings.TrimSpace(GetRegisteredToolAppendixLocale(BuiltinToolSearch.Code, locale)); appendix != "" {
|
||||
ret = append(ret, appendix)
|
||||
}
|
||||
}
|
||||
for _, spec := range RegisteredToolSpecs {
|
||||
if strings.TrimSpace(spec.Appendix) == "" {
|
||||
continue
|
||||
}
|
||||
if spec.Code == BuiltinToolSearch.Code {
|
||||
continue
|
||||
}
|
||||
if containsNormalizedToolCode(normalizedToolCodes, spec.Code) {
|
||||
ret = append(ret, spec.Appendix)
|
||||
if appendix := strings.TrimSpace(GetRegisteredToolAppendixLocale(spec.Code, locale)); appendix != "" {
|
||||
ret = append(ret, appendix)
|
||||
}
|
||||
}
|
||||
}
|
||||
return ret
|
||||
|
||||
@@ -53,12 +53,12 @@ func TestRegisteredToolTextUsesEnglishLocale(t *testing.T) {
|
||||
|
||||
func TestRegisteredToolTextKeepsChineseLocale(t *testing.T) {
|
||||
title := GetRegisteredToolTitleLocale(GraphCreateTicketConfirm.Code, i18nx.LocaleZhCN)
|
||||
if title != GraphCreateTicketConfirm.Title {
|
||||
if title != "创建工单确认流程" {
|
||||
t.Fatalf("unexpected chinese title: %q", title)
|
||||
}
|
||||
|
||||
description := GetRegisteredToolDescriptionLocale(GraphCreateTicketConfirm.Code, i18nx.LocaleZhCN)
|
||||
if description != GraphCreateTicketConfirm.Description {
|
||||
if description != "Graph Tool。处理工单参数准备、用户确认、实际创建工单和结果返回。" {
|
||||
t.Fatalf("unexpected chinese description: %q", description)
|
||||
}
|
||||
}
|
||||
|
||||
Reference in New Issue
Block a user