تحويل الدعم من الخط الأول
يقرأ أسئلة العملاء الواردة، ويجاوب تلقائياً على أكثرها شيوعاً بنسبة 60-70% باستخدام ردود مبنية على محتوى معرفة العميل، ويحوّل أي شي غير مؤكد لإنسان مع إرفاق السياق مسبقاً. العميل يحس بدعم أسرع، وطابور فريق الدعم يظل مركّز على الأسئلة الصعبة فعلاً، والإجابات تبقى دقيقة لأن النمط يبني كل رد على وثائق العميل الحقيقية مو على بيانات تدريبه.
المتطلبات تصف قدرات يحتاجها النمط في بيئتك، مو الموردين اللي لازم تشتريهم. أي نظام يملأ متطلباً يحقّقه — وهذا اللي يخلّي الكتالوج قابلاً للنقل عبر الذيل الطويل من أدوات الشركات الصغيرة والمتوسطة.
inbound_request_channelThe channel where customers send their questions and the pattern listens for new ones.
- ticket created in a helpdesk system
- message arrives in a shared support inbox
- chat session opened in a website widget
- form submission from a contact page
response_channelThe channel through which replies reach the customer. Often but not always the same system as the inbound channel.
- reply posted into the same helpdesk ticket
- email sent from a shared inbox
- message returned in the chat session
knowledge_corpusThe factual content the pattern grounds answers on. Without this, the pattern hallucinates and can't be trusted in production.
- internal wiki or knowledge base
- published help center articles
- collection of policy documents in a file store
- product documentation site
human_escalation_routeWhere uncertain or low-confidence cases go so a human can take over without losing context.
- ticket reassigned within the helpdesk with internal notes
- alert posted to a team chat channel with a deep link
- email forwarded to a senior agent inbox
customer_context_storeLookup of who is asking and their prior history with the business. Helps the pattern personalize and avoid contradicting prior decisions.
- CRM with customer profiles and interaction history
- e-commerce platform with order history
- subscription billing system with account state
- internal customer database
supervisor_review_queuePeriodic sample of pattern-handled cases that a human reviews to catch quality drift over time.
- weekly digest emailed to a designated reviewer
- dashboard with sampled interactions
- ticket queue tagged 'AI-review-required' in the helpdesk
- 01Listen for a new inquiry on the inbound channel and capture its content and sender identifier
inbound_request_channel - 02If sender identifier is recognized, fetch customer context to personalize the response
customer_context_storeقرار Skip this step if customer_context_store is not filled or sender is anonymous. - 03Classify the inquiry intent against the pattern's working taxonomy of 10-20 common topics tuned per client
- 04Retrieve relevant passages from the knowledge corpus using semantic search constrained by detected intent
knowledge_corpus - 05Generate a candidate response grounded on retrieved passages, with confidence score and citations
- 06Evaluate confidence and content against guardrails: minimum confidence threshold, no advice in regulated areas, no commitments on price or policy without sourceقرار If confidence ≥ threshold AND guardrails pass, proceed to step 7. Otherwise jump to step 8.
- 07Post the response to the response channel as the configured responder identity, log the interaction for review sampling
response_channelsupervisor_review_queue - 08Route to a human agent with the original inquiry, draft (if any), reasoning, confidence score, and suggested owner
human_escalation_route
مخرجات منظّمة ينتجها هذا النمط. أنماط ثانية وأنظمة العملاء تقدر تشترك فيها، وهكذا يتركّب الكتالوج مع الوقت.
labeled_inquiryEach customer inquiry classified by intent against the pattern's taxonomy.
- analytics dashboards tracking topic trends
- product teams looking for friction signals
- knowledge base maintenance workflows that surface frequently-asked questions missing from docs
escalation_eventEach escalation to a human, with the reason.
- support team capacity planning
- later patterns that may handle the escalated category
- training feedback loops to expand the pattern's coverage
review_sampleSampled responses queued for human supervisor review, with the reviewer's eventual verdict.
- quality dashboards
- training data for prompt refinement and eval set growth