{"id":"44ac31d2-0e45-4164-a9e1-9b59d52ee6b3","entityType":"agent","slug":"clawhub-leooooooow-chatbot-designer","name":"Chatbot Designer","canonicalUrl":"https://www.xpersona.co/agent/clawhub-leooooooow-chatbot-designer","canonicalPath":"/agent/clawhub-leooooooow-chatbot-designer","generatedAt":"2026-10-11T17:43:01.782Z","source":"CLAWHUB","claimStatus":"UNCLAIMED","verificationTier":"NONE","summary":{"evidence":{"source":"editorial-content","verified":true,"confidence":"high","updatedAt":"2026-10-11T15:35:57.838Z","emptyReason":null},"description":"Design customer service chatbot conversation flows for ecommerce — order status, returns, product recommendations, and escalation rules — that reduce ticket... Skill: Chatbot Designer Owner: leooooooow Summary: Design customer service chatbot conversation flows for ecommerce — order status, returns, product recommendations, and escalation rules — that reduce ticket... Tags: latest:1.1.0 Version history: v1.1.0 | 2026-06-05T14:44:21.110Z | user **Changelog – Version 1.0.1** - Added four documentation files: escalation-playbook.md, intent-library.md, output-template.md, and q","descriptionLabel":"Technical summary","evidenceSummary":"Capability contract not published. No trust telemetry is available yet. 1K downloads reported by the source. Last updated 10/11/2026.","installCommand":"clawhub skill install s17974h9acjg4h7h5djv1hg51d83hcca:chatbot-designer","sourceUrl":"https://clawhub.ai/leooooooow/chatbot-designer","homepage":"https://clawhub.ai/leooooooow/skills/chatbot-designer","primaryLinks":[{"label":"View on ClawHub","url":"https://clawhub.ai/leooooooow/chatbot-designer","kind":"source"},{"label":"Homepage","url":"https://clawhub.ai/leooooooow/skills/chatbot-designer","kind":"homepage"}],"safetyScore":84,"overallRank":62,"popularityScore":60,"trustScore":null,"claimedByName":null,"isOwner":false,"seoDescription":"Design customer service chatbot conversation flows for ecommerce — order status, returns, product recommendations, and escalation rules — that reduce ticket... "},"coverage":{"evidence":{"source":"public-profile","verified":false,"confidence":"medium","updatedAt":"2026-10-11T15:35:57.838Z","emptyReason":null},"protocols":[{"protocol":"OPENCLEW","label":"OpenClaw","status":"self-declared","notes":"Declared in the public agent profile."}],"capabilities":[],"verifiedCount":0,"selfDeclaredCount":1,"capabilityMatrix":{"rows":[{"key":"OPENCLEW","type":"protocol","support":"unknown","confidenceSource":"profile","notes":"Listed on profile"}],"flattenedTokens":"protocol:OPENCLEW|unknown|profile"}},"adoption":{"evidence":{"source":"CLAWHUB","verified":false,"confidence":"medium","updatedAt":"2026-10-11T15:35:57.838Z","emptyReason":null},"stars":null,"forks":null,"downloads":1038,"packageName":null,"latestVersion":"1.1.0","tractionLabel":"1K downloads"},"release":{"evidence":{"source":"CLAWHUB","verified":false,"confidence":"medium","updatedAt":"2026-10-11T15:35:57.824Z","emptyReason":null},"lastUpdatedAt":"2026-10-11T15:35:57.838Z","lastCrawledAt":"2026-10-11T15:35:57.824Z","lastIndexedAt":null,"nextCrawlAt":"2026-10-12T15:35:57.824Z","lastVerifiedAt":null,"highlights":[{"version":"1.1.0","createdAt":"2026-06-05T14:44:21.110Z","changelog":"**Changelog – Version 1.0.1** - Added four documentation files: escalation-playbook.md, intent-library.md, output-template.md, and quality-checklist.md to improve skill guidance and structure. - Removed outdated file: skill-card.md. - Expanded SKILL.md with a comprehensive quick reference, step-by-step workflow, and practical decision guidelines. - Enhanced details on intent mapping, escalation protocols, conversation flow design, and response writing. - Clarified outputs and best practices for building effective ecommerce customer service chatbots.","fileCount":7,"zipByteSize":15216},{"version":"1.0.0","createdAt":"2026-04-17T01:08:53.546Z","changelog":"Initial release.","fileCount":3,"zipByteSize":4352}]},"execution":{"evidence":{"source":"CLAWHUB","verified":false,"confidence":"low","updatedAt":null,"emptyReason":"No published capability contract is available yet."},"installCommand":"clawhub skill install s17974h9acjg4h7h5djv1hg51d83hcca:chatbot-designer","setupComplexity":"low","setupSteps":["Setup complexity is LOW. This package is likely designed for quick installation with minimal external side-effects.","Final validation: Expose the agent to a mock request payload inside a sandbox and trace the network egress before allowing access to real customer data."],"contract":{"contractStatus":"missing","authModes":[],"requires":[],"forbidden":[],"supportsMcp":false,"supportsA2a":false,"supportsStreaming":false,"inputSchemaRef":null,"outputSchemaRef":null,"dataRegion":null,"contractUpdatedAt":null,"sourceUpdatedAt":null,"freshnessSeconds":null},"invocationGuide":{"preferredApi":{"snapshotUrl":"https://www.xpersona.co/api/v1/agents/clawhub-leooooooow-chatbot-designer/snapshot","contractUrl":"https://www.xpersona.co/api/v1/agents/clawhub-leooooooow-chatbot-designer/contract","trustUrl":"https://www.xpersona.co/api/v1/agents/clawhub-leooooooow-chatbot-designer/trust"},"curlExamples":["curl -s \"https://www.xpersona.co/api/v1/agents/clawhub-leooooooow-chatbot-designer/snapshot\"","curl -s \"https://www.xpersona.co/api/v1/agents/clawhub-leooooooow-chatbot-designer/contract\"","curl -s \"https://www.xpersona.co/api/v1/agents/clawhub-leooooooow-chatbot-designer/trust\""],"jsonRequestTemplate":{"query":"summarize this repo","constraints":{"maxLatencyMs":2000,"protocolPreference":["OPENCLEW"]}},"jsonResponseTemplate":{"ok":true,"result":{"summary":"...","confidence":0.9},"meta":{"source":"CLAWHUB","generatedAt":"2026-10-11T17:43:01.780Z"}},"retryPolicy":{"maxAttempts":3,"backoffMs":[500,1500,3500],"retryableConditions":["HTTP_429","HTTP_503","NETWORK_TIMEOUT"]}},"endpoints":{"dossierUrl":"https://www.xpersona.co/api/v1/agents/clawhub-leooooooow-chatbot-designer/dossier","snapshotUrl":"https://www.xpersona.co/api/v1/agents/clawhub-leooooooow-chatbot-designer/snapshot","contractUrl":"https://www.xpersona.co/api/v1/agents/clawhub-leooooooow-chatbot-designer/contract","trustUrl":"https://www.xpersona.co/api/v1/agents/clawhub-leooooooow-chatbot-designer/trust"}},"reliability":{"evidence":{"source":"runtime-metrics","verified":false,"confidence":"low","updatedAt":null,"emptyReason":"No trust, reliability, or runtime telemetry is available."},"trust":{"status":"unavailable","handshakeStatus":"UNKNOWN","verificationFreshnessHours":null,"reputationScore":null,"p95LatencyMs":null,"successRate30d":null,"fallbackRate":null,"attempts30d":null,"trustUpdatedAt":null,"trustConfidence":"unknown","sourceUpdatedAt":null,"freshnessSeconds":null},"decisionGuardrails":{"doNotUseIf":["Contract metadata is missing or unavailable for deterministic execution."],"safeUseWhen":[],"riskFlags":["missing_or_unavailable_contract","trust_data_unavailable","schema_references_missing"],"operationalConfidence":"low"},"executionMetrics":{"observedLatencyMsP50":null,"observedLatencyMsP95":null,"estimatedCostUsd":null,"uptime30d":null,"rateLimitRpm":null,"rateLimitBurst":null,"lastVerifiedAt":null,"verificationSource":null},"runtimeMetrics":{"successRate":null,"avgLatencyMs":null,"avgCostUsd":null,"hallucinationRate":null,"retryRate":null,"disputeRate":null,"p50Latency":null,"p95Latency":null,"lastUpdated":null}},"benchmarks":{"evidence":{"source":"no-benchmark-data","verified":false,"confidence":"low","updatedAt":null,"emptyReason":"No benchmark suites or observed failure patterns are available."},"suites":[],"failurePatterns":[]},"artifacts":{"evidence":{"source":"CLAWHUB","verified":false,"confidence":"high","updatedAt":"2026-10-11T15:35:57.838Z","emptyReason":null},"readme":"Skill: Chatbot Designer\n\nOwner: leooooooow\n\nSummary: Design customer service chatbot conversation flows for ecommerce — order status, returns, product recommendations, and escalation rules — that reduce ticket...\n\nTags: latest:1.1.0\n\nVersion history:\n\nv1.1.0 | 2026-06-05T14:44:21.110Z | user\n\n**Changelog – Version 1.0.1**\n\n- Added four documentation files: escalation-playbook.md, intent-library.md, output-template.md, and quality-checklist.md to improve skill guidance and structure.\n- Removed outdated file: skill-card.md.\n- Expanded SKILL.md with a comprehensive quick reference, step-by-step workflow, and practical decision guidelines.\n- Enhanced details on intent mapping, escalation protocols, conversation flow design, and response writing.\n- Clarified outputs and best practices for building effective ecommerce customer service chatbots.\n\nv1.0.0 | 2026-04-17T01:08:53.546Z | user\n\nInitial release.\n\nArchive index:\n\nArchive v1.1.0: 7 files, 15216 bytes\n\nFiles: escalation-playbook.md (4880b), intent-library.md (5704b), output-template.md (1978b), quality-checklist.md (2717b), skill-card.md (2644b), SKILL.md (13306b), _meta.json (135b)\n\nFile v1.1.0:SKILL.md\n\n---\nname: Chatbot Designer\ndescription: Design customer service chatbot conversation flows for ecommerce — order status, returns, product recommendations, and escalation rules — that reduce ticket volume while maintaining satisfaction scores.\n---\n\n# Chatbot Designer\n\nDesign customer service chatbot conversation flows for ecommerce including order status inquiries, return requests, product recommendations, and escalation rules that reduce ticket volume while maintaining satisfaction scores. Most chatbot failures come not from the technology but from poorly designed flows — dead ends, missing escalation paths, or responses that feel robotic. This skill produces conversation architecture, intent mapping, and response logic that resolves common queries automatically while seamlessly handing off complex issues to human agents.\n\n## Quick Reference\n\n| Decision | Strong | Acceptable | Weak |\n|---|---|---|---|\n| Intent coverage | Map top 10 intents from ticket data before writing any flows | Map intents from assumption | Build flows for every possible scenario before launching |\n| Escalation design | Graceful escalation after 2 failed responses + any customer frustration signal | Escalation only when explicitly requested | No escalation path — bot loops or dead-ends |\n| Tone | Warm, direct, brand-consistent; acknowledges frustration | Neutral corporate tone | Overly casual/emoji-heavy OR stiff formal robospeak |\n| Resolution path | Bot resolves or creates ticket — never leaves customer waiting without next step | Resolves common issues, drops others | Gives information without action (\"call this number\") |\n| Order integration | Connected to OMS so bot can pull live order status | Static FAQs about shipping timelines | Cannot access order data at all |\n| Return flow | Initiates return label automatically; confirms eligibility in real time | Explains return policy; directs to email | Can't process returns; sends customer elsewhere |\n| Measurement | CSAT on bot interactions + containment rate + escalation rate tracked weekly | Track containment rate only | No metrics |\n\n## Solves\n\n- Support team overwhelmed with high-volume, repetitive queries (order status, return requests, shipping ETAs)\n- Long first-response times hurting customer satisfaction and review scores\n- After-hours support gaps — customers getting no response outside business hours\n- Support costs scaling linearly with order volume instead of leveling off\n- Inconsistent answers from human agents on standard policy questions\n- High ticket volume from customers who can't self-serve on the website\n- Returns and exchanges handled manually when they could be automated end-to-end\n\n## Workflow\n\n### Step 1 — Map Your Top 10 Support Intents\n\nPull 90 days of support ticket data and categorize by intent. Most ecommerce stores find 80% of volume concentrated in 5–7 intents.\n\n**Typical ecommerce intent distribution:**\n\n| Intent | Average % of tickets | Automatable? |\n|---|---|---|\n| Order status / tracking | 28–35% | ✅ Fully (with OMS integration) |\n| Return / exchange request | 18–24% | ✅ Mostly (policy check + label generation) |\n| Shipping delay inquiry | 10–15% | ✅ Partially (status pull + proactive message) |\n| Product question (pre-purchase) | 8–12% | ✅ Partially (FAQ lookup) |\n| Wrong item received | 5–8% | ✅ Partial (initiate replacement flow) |\n| Payment / billing issue | 4–7% | ⚠️ Partial (info only; resolution needs human) |\n| Discount / promo code issue | 3–5% | ✅ Mostly (validate code, explain terms) |\n| Cancellation request | 3–5% | ✅ If pre-fulfillment; ⚠️ if shipped |\n| Product defect / damage | 3–5% | ✅ Initiate claim; photo required |\n| General feedback | 2–4% | ⚠️ Log and escalate |\n\nPrioritize the top 3–5 intents for chatbot coverage at launch. Build 100% coverage of these before adding lower-volume intents.\n\n### Step 2 — Design the Main Menu Architecture\n\nEvery chatbot needs a clear entry point. Design the main menu to match your top intents — never more than 5–6 options.\n\n**Recommended main menu structure:**\n\n```\nHi [name]! How can I help today?\n  → 📦 Track my order\n  → 🔄 Return or exchange\n  → ❓ Product question\n  → 💳 Billing or payment\n  → 🙋 Talk to a person\n```\n\nRules:\n- Always include \"Talk to a person\" — hiding this frustrates customers and tanks CSAT\n- Use icons for scan-ability on mobile\n- Keep labels under 4 words\n- Offer free-text input alongside menu (some customers prefer to type)\n\n### Step 3 — Design Each Intent Flow\n\nFor each priority intent, design a complete flow with: entry points, required data collection, decision branches, resolution, and escalation exit.\n\n**Flow template structure:**\n\n```\n1. TRIGGER: Intent detected (menu selection or keyword match)\n2. ACKNOWLEDGE: \"Let me pull that up for you.\"\n3. DATA COLLECT: What does the bot need? (order number, email, product name)\n4. LOOKUP / CHECK: Connect to data source or apply policy rules\n5. RESOLUTION BRANCH A: Can resolve → confirm resolution, offer next step\n6. RESOLUTION BRANCH B: Cannot resolve → explain why + escalate gracefully\n7. ESCALATION: Create ticket with full context pre-filled; set expectation (\"Team will reply in 2–4 hours\")\n8. CONFIRMATION: Always confirm what happened before ending conversation\n```\n\n**Order Status Flow example:**\n\n```\nBot: \"To track your order, I'll need your order number or the email you used to order.\"\nCustomer: [provides order #12345]\nBot: [looks up OMS] → \"Order #12345 was shipped on June 3rd via FedEx.\n     Estimated delivery: June 7th.\n     Tracking: [link]\n     Is there anything else you'd like to know about this order?\"\n     → Yes → return to order menu\n     → No → \"Thanks! If your package doesn't arrive by June 8th, come back and I'll help you file a claim.\"\n```\n\n### Step 4 — Design the Return / Exchange Flow\n\nReturns are the highest-value flow to automate — they're high-volume, rule-based, and automatable.\n\n**Return eligibility check decision tree:**\n\n```\nCustomer: \"I want to return my order\"\nBot: \"I can help with that. What's your order number?\"\n→ [Order number lookup]\n→ Check: Is order within return window? (e.g., 30 days from delivery)\n   → YES: \"Great — this order is eligible. What's the reason for your return?\"\n           → [Reason menu: wrong size / doesn't meet expectations / damaged / wrong item]\n           → [Generate return label or RMA number]\n           → \"Your prepaid return label has been sent to [email]. Please drop off within 7 days.\"\n   → NO: \"This order was delivered on [date] — our 30-day window closed on [date].\n           I'm not able to process this automatically, but I'll connect you with our team\n           who can review exceptions.\"\n           → [Create ticket: late return request, order details pre-filled]\n```\n\n### Step 5 — Design Escalation Protocols\n\nEscalation design is where most chatbots fail. Build these rules:\n\n**Automatic escalation triggers:**\n- Customer uses frustration signals: \"this is ridiculous,\" \"never again,\" \"unacceptable,\" \"worst,\" \"furious\"\n- Bot fails to resolve the same intent after 2 attempts\n- Customer types \"human,\" \"agent,\" \"real person,\" \"help me\"\n- Issue involves fraud, legal threat, or physical safety\n- High-value customer (flag via CRM/LTV tag) — route to priority queue\n\n**Escalation experience requirements:**\n1. Never apologize for failing — transition warmly: \"Let me get someone on our team to help with this.\"\n2. Pre-fill the ticket with full conversation transcript so customer doesn't repeat themselves\n3. Set a specific response time expectation: \"Our team will reply by [time] — typically within 2–4 business hours.\"\n4. Offer async confirmation: \"You'll get a notification at [email] when we respond.\"\n\n**Escalation quality test:**\nGo through every flow yourself and try to trigger frustration or dead-ends. If you can get the bot stuck or leave without a resolution, fix it before launch.\n\n### Step 6 — Write Response Copy\n\nBot copy must feel human without pretending to be human. Guidelines:\n\n**Tone principles:**\n- Acknowledge before acting: \"I'll take a look at that for you\" not \"Order number required.\"\n- Use contractions: \"I'll\" not \"I will\"; \"you're\" not \"you are\"\n- Be specific: \"Your order ships in 3–5 business days\" not \"orders ship soon\"\n- Validate frustration without over-apologizing: \"I understand that's frustrating — let me help fix it\" not \"I'm SO sorry for this terrible experience!!!!\"\n- Avoid robotic completions: \"Is there anything else I can help you with today?\" → \"Anything else on your mind?\" is warmer\n\n**Response length:**\n- Confirmations: 1–2 sentences\n- Policy explanations: 3–4 sentences max, then offer human if still confused\n- Error messages: Always include what to do next, never just what failed\n\n### Step 7 — Measure and Iterate\n\n**Core chatbot KPIs:**\n\n| Metric | Definition | Target |\n|---|---|---|\n| Containment rate | % of conversations bot resolves without human escalation | >60% at 3 months |\n| Escalation rate | % of conversations handed to human | <30% |\n| CSAT (bot interactions) | Customer satisfaction score post-bot conversation | >3.8 / 5 |\n| First contact resolution | % of issues resolved in the first interaction | >50% |\n| Drop-off rate | % leaving conversation without resolution or escalation | <15% |\n| Top unhandled intents | Intents with no matching flow | Review weekly |\n\nReview unhandled intent logs weekly for the first 3 months — these are your next flows to build.\n\n## Examples\n\n### Example 1 — Fashion Brand (Shopify + Gorgias Chatbot)\n\n**Setup:** 450 support tickets/week; top intents: order status (34%), returns (21%), sizing questions (14%)  \n**Tool:** Gorgias with Shopify integration  \n**Flows built at launch:** Order tracking, return initiation, size guide lookup\n\n**Order tracking flow performance:**\n- 89% of order status queries resolved by bot without human\n- Average resolution time: 8 seconds (vs. 4-hour human response time)\n- CSAT on bot interactions: 4.1/5\n\n**Return flow:**\n- 74% of eligible returns processed end-to-end by bot\n- Average time to label generation: 2 minutes\n- Support ticket volume down 31% within 60 days of launch\n\n---\n\n### Example 2 — Electronics Accessories (WooCommerce + Tidio)\n\n**Challenge:** High volume of pre-purchase compatibility questions that required manual lookup  \n**Solution:** Product compatibility decision tree built in Tidio, connected to product specs database\n\n**Flow:**\n```\n\"What device are you trying to use this with?\"\n→ [Device selection menu: iPhone 15 / iPhone 14 / Samsung S24 / Other]\n→ \"The [product] is fully compatible with [device]. It includes [cable type] in the box.\n   Want to add it to your cart?\"\n```\n\n**Result:**\n- Pre-purchase conversion increased 8% (buying confidence from instant compatibility confirmation)\n- Compatibility-related support tickets down 44%\n- Human agents redirected from answering repetitive compatibility questions to handling escalations and complex issues\n\n## Common Mistakes\n\n1. **Building flows from assumption instead of ticket data** — Designing 20 flows for hypothetical scenarios while missing the 3 intents that make up 60% of actual ticket volume.\n\n2. **No escalation path** — The single most damaging mistake. A bot that loops or dead-ends when it can't resolve an issue destroys trust faster than no bot at all.\n\n3. **Pretending to be human** — Customers who discover they're talking to a bot after believing it was a person feel deceived. Be transparent: \"Hi! I'm [BrandName]'s virtual assistant.\"\n\n4. **Copy-pasting canned responses** — Bot responses that sound like they came from a 2005 helpdesk FAQ kill the experience. Write for conversation, not documentation.\n\n5. **Launching without OMS integration** — A chatbot that can't actually look up order data can only answer generic shipping questions, not the specific question the customer has.\n\n6. **Ignoring mobile UX** — Most customer service chatbot interactions happen on mobile. Long menus, small tap targets, and text-heavy responses that require scrolling kill completion rates.\n\n7. **No metrics review schedule** — Bots built and forgotten miss the unhandled intents that accumulate weekly. Review intent logs every week for the first 3 months.\n\n8. **Over-qualifying before helping** — Asking for order number, email, AND phone number before providing any assistance feels like an interrogation. Collect the minimum needed for each specific flow.\n\n9. **CSAT survey fatigue** — Sending a satisfaction survey after every bot interaction annoys customers. Survey 20–30% of interactions randomly, or only after escalation.\n\n10. **Trying to handle everything at launch** — 5 well-designed flows that cover 70% of volume outperform 20 incomplete flows covering 100% of scenarios with gaps and dead ends.\n\n## Resources\n\n- [Output Template](references/output-template.md) — Chatbot flow design document\n- [Intent Library](references/intent-library.md) — Pre-built intent patterns and response templates\n- [Escalation Playbook](references/escalation-playbook.md) — Escalation trigger rules and handoff scripts\n- [Quality Checklist](assets/quality-checklist.md) — Pre-launch chatbot review checklist\n\nFile v1.1.0:_meta.json\n\n{\n  \"ownerId\": \"kn70fv0ehp50emedet9tx3fekd82pw3b\",\n  \"slug\": \"chatbot-designer\",\n  \"version\": \"1.1.0\",\n  \"publishedAt\": 1780670661110\n}\n\nFile v1.1.0:escalation-playbook.md\n\n# Escalation Playbook\n\n## Escalation Philosophy\n\nAn escalation is not a failure — it is the chatbot doing its job. The goal is to:\n1. Catch the moment when a human is needed\n2. Transfer context so the customer doesn't repeat themselves\n3. Set an accurate expectation for response time\n4. Leave the customer feeling helped, not abandoned\n\nA chatbot with excellent escalation design achieves higher CSAT than one that tries to resolve everything and gets stuck.\n\n## Escalation Trigger Taxonomy\n\n### Tier 1 — Immediate Escalation (within seconds)\n\n| Trigger | Detection method | Queue |\n|---|---|---|\n| Physical safety mention | Keywords: injury, hurt, damaged, burned, allergic reaction | Emergency / Senior |\n| Legal or fraud mention | Keywords: fraud, dispute, chargeback, lawyer, report | Legal / Senior |\n| High-value customer | CRM tag: LTV > $[threshold] | Priority |\n| Explicit human request | Keywords: human, agent, person, real | General |\n| Severe frustration language | Keywords: furious, unacceptable, disgusting, lawsuit | Priority |\n\n### Tier 2 — Escalation After 2 Failed Attempts\n\n| Trigger | Condition | Queue |\n|---|---|---|\n| Unresolvable intent | Bot attempted twice, no resolution | General |\n| Missing order data | Cannot find order after 2 lookups | General |\n| Out-of-policy request | Outside return window, discontinued product | General |\n| Complex multi-issue | Customer raises 3+ different issues in one session | General |\n\n### Tier 3 — Proactive Escalation (Batched)\n\n| Trigger | Timing | Action |\n|---|---|---|\n| Unhandled intent | Weekly review | Add to flow build queue |\n| Low CSAT bot rating | <3/5 rating given | Review transcript; identify gap |\n| Long session with no resolution | Session > 10 min, no confirmation | Flag for quality review |\n\n## Escalation Handoff Protocol\n\n### Step 1 — Signal the transition warmly\nNever abruptly transfer. Use a bridging line:\n> \"Let me get one of our team members to help you with this.\"\n> \"This is something I want to make sure our specialists handle for you.\"\n> \"I'm connecting you now — you're in good hands.\"\n\n### Step 2 — Summarize the conversation\nPre-fill the ticket with:\n- Customer name and contact\n- Order number(s) involved\n- Issue summary in 2–3 sentences\n- What the bot already tried\n- Customer sentiment (frustrated / neutral / positive)\n- Any data already collected (return reason, device type, etc.)\n\n**Ticket pre-fill template:**\n```\n[BOT HANDOFF] Customer: [Name] | Order: #[number]\nIssue: [1-sentence summary]\nBot actions taken: [what was attempted]\nData collected: [relevant fields]\nCustomer sentiment: [frustrated/neutral/satisfied]\nChat transcript: [attached]\n```\n\n### Step 3 — Set response time expectation\nNever use vague language. Be specific:\n- ✅ \"Our team will reply within 2 business hours\"\n- ✅ \"You'll hear back by 9am tomorrow\"\n- ❌ \"Someone will be in touch soon\"\n- ❌ \"We'll reply as quickly as possible\"\n\n### Step 4 — Confirm and close the bot interaction\n> \"I've sent everything to our team and you'll receive a reply at [email] by [time]. Your reference number is #[ticket]. Is there anything else I can help with before I hand off?\"\n\n## Queue Routing Logic\n\n| Queue | Criteria | Target SLA |\n|---|---|---|\n| Emergency | Safety, legal, fraud | 30 minutes |\n| Priority | LTV > threshold, severe frustration, escalation from chatbot failure | 1 business hour |\n| General | Standard escalations | 2–4 business hours |\n| After-hours | Outside business hours | By 9am next business day |\n\n## After-Hours Handling\n\nWhen escalating outside business hours:\n1. Acknowledge the timing: \"Our team is offline right now (back at [time]).\"\n2. Confirm the ticket is created with a reference number\n3. Give a specific \"by when\" for response — not just \"tomorrow\"\n4. Offer any self-serve options that might help in the meantime\n\n**After-hours script:**\n> \"Our team is offline right now — we're back at [9am time zone]. I've created a priority ticket for you (#[number]) and someone will reply to [email] by [specific time]. In the meantime, you can check [tracking link / FAQ link] for [relevant self-serve option].\"\n\n## CSAT Recovery Protocol\n\nWhen a customer rates their bot interaction <3/5:\n1. Auto-trigger human follow-up within 4 hours\n2. Human reviews transcript before reaching out\n3. Proactive resolution offer: \"I noticed your experience wasn't great — I'd like to make it right.\"\n4. Log the transcript in the quality review queue for flow improvement\n\n## Escalation Metrics to Track\n\n| Metric | Target | Action if off-target |\n|---|---|---|\n| Escalation rate | <30% | Review top escalated intents; build/fix flows |\n| Escalation CSAT | >4.0/5 | Review handoff scripts and response time compliance |\n| SLA compliance | >90% | Staffing or routing adjustment |\n| Repeat contacts | <10% of escalated cases | Post-resolution follow-up improving |\n\nFile v1.1.0:intent-library.md\n\n# Intent Library — Ecommerce Chatbot\n\n## Order Status Intent\n\n**Trigger keywords:** \"where is my order\", \"track\", \"tracking\", \"shipping status\", \"hasn't arrived\", \"order status\", \"when will it arrive\", \"delivery\"\n\n**Required data:** Order number OR email address\n\n**Success response template:**\n> \"I found your order! Order #[number] shipped on [date] via [carrier]. Your tracking number is [tracking] — click here to see live updates: [link]. Estimated delivery: [date]. Is there anything else I can help with?\"\n\n**Delayed order response:**\n> \"Your order #[number] is currently showing a delay with [carrier]. The new estimated delivery is [date]. I've flagged this on your account. If it doesn't arrive by [date+1], come back and I can file a claim on your behalf.\"\n\n**Not found response:**\n> \"I couldn't find an order with that information. Can you double-check the order number? It's usually in your confirmation email and starts with #. If you still can't find it, let me connect you with our team.\"\n\n---\n\n## Return Request Intent\n\n**Trigger keywords:** \"return\", \"refund\", \"exchange\", \"send back\", \"doesn't fit\", \"changed my mind\", \"not what I expected\"\n\n**Required data:** Order number, reason for return\n\n**Eligible return response:**\n> \"I can process this return for you. Your order is within our [30-day] return window. I'll send a prepaid return label to [email] — please allow 5–10 minutes for it to arrive. Once we receive your return, your refund will be processed within [X] business days.\"\n\n**Ineligible return response (outside window):**\n> \"This order was delivered on [date] — our [30-day] return window closed on [date]. I'm not able to process this automatically, but I'll connect you with our team who can review your situation.\"\n\n**Exchange response:**\n> \"For an exchange, I'll need to process a return and place a new order. Would you like me to start the return process? You can then place a new order with the correct [size/color/item], and I can apply a discount code for the inconvenience.\"\n\n---\n\n## Shipping Delay Intent\n\n**Trigger keywords:** \"late\", \"delayed\", \"hasn't shipped\", \"still processing\", \"stuck\", \"where\", \"overdue\"\n\n**Response template:**\n> \"I'm sorry to hear your order is running late. Let me check on this. [Look up order status] Your order is currently [status]. [If in transit:] It looks like there's been a carrier delay — the new estimated delivery is [date]. [If not yet shipped:] It's still being prepared for shipment. I'll flag this as high priority for our fulfillment team.\"\n\n---\n\n## Wrong Item Received Intent\n\n**Trigger keywords:** \"wrong item\", \"wrong product\", \"not what I ordered\", \"incorrect\", \"different item\"\n\n**Response template:**\n> \"I'm really sorry about that — receiving the wrong item is frustrating. I'd like to make this right. Can you share: (1) your order number, and (2) what you received vs. what you ordered? I'll initiate a replacement shipment and arrange return of the incorrect item at no cost to you.\"\n\n---\n\n## Product Question Intent (Pre-Purchase)\n\n**Trigger keywords:** \"compatible\", \"does it work with\", \"what size\", \"fit\", \"material\", \"dimensions\", \"specifications\"\n\n**Response template:**\n> \"Great question! [Answer from FAQ database]. [If not found:] I don't have that specific detail on hand — let me connect you with someone who can confirm before you order.\"\n\n**Compatibility check template:**\n> \"The [product] is [compatible/not compatible] with [device/model]. [Add specific detail.] Does that help, or would you like to check something else?\"\n\n---\n\n## Cancellation Request Intent\n\n**Trigger keywords:** \"cancel\", \"don't want it\", \"change my mind\", \"cancel order\"\n\n**Pre-shipment response:**\n> \"I can cancel this order since it hasn't shipped yet. Confirming cancellation of order #[number] — your refund of $[amount] will appear within 3–5 business days. Is there anything I can help you find instead?\"\n\n**Post-shipment response:**\n> \"Unfortunately, order #[number] has already shipped, so I'm not able to cancel it at this stage. Once you receive it, I can process a return with a free prepaid label. Would you like me to set that up in advance?\"\n\n---\n\n## Promo Code Issue Intent\n\n**Trigger keywords:** \"code doesn't work\", \"discount not applying\", \"coupon\", \"promo\", \"won't accept\"\n\n**Response template:**\n> \"Let me check on that code. [Validate code status] [If valid but not applying:] This code applies to [eligible products/categories] — it looks like your cart may have [ineligible item]. Try removing [item] and re-entering the code. [If expired:] This code expired on [date]. Would you like me to check if there's a current offer I can apply? [If not found:] I couldn't find this code in our system. Can you double-check the spelling? Promo codes are case-sensitive.\"\n\n---\n\n## Escalation Handoff Script Templates\n\n**Warm handoff (frustration detected):**\n> \"I can see this has been really frustrating, and I want to make sure you get the best help possible. I'm connecting you now with [Name/Team] — they specialize in situations like yours. I've shared our entire conversation so you won't need to repeat anything. They'll reply to you at [email] within [timeframe].\"\n\n**After-hours escalation:**\n> \"Our team is currently offline, but I've created a priority ticket for you (#[ticket number]). You'll receive a reply at [email] by [specific time] tomorrow. In the meantime, here's what I've noted about your issue: [summary].\"\n\n**High-value customer escalation:**\n> \"You're one of our valued customers, and I want to make sure this is handled by our senior support team. I'm flagging this as priority and someone will be in touch within [1–2 hours] during business hours.\"\n\nFile v1.1.0:output-template.md\n\n# Chatbot Flow Design Document\n\n## Project Overview\n- **Brand:** \n- **Platform:** Shopify / WooCommerce / Custom\n- **Chatbot tool:** Gorgias / Tidio / Intercom / Zendesk / Custom\n- **OMS integration:** Yes / No / Planned\n- **Launch date target:** \n\n## Intent Priority Matrix\n| Rank | Intent | % of ticket volume | Automation potential | Build priority |\n|---|---|---|---|---|\n| 1 | | % | High/Medium/Low | Launch / Phase 2 |\n| 2 | | % | | |\n| 3 | | % | | |\n| 4 | | % | | |\n| 5 | | % | | |\n\n## Main Menu Design\n```\nWelcome message: \n\nMenu options:\n1. \n2. \n3. \n4. \n5. Talk to a person\n```\n\n## Flow Designs\n\n### Flow 1: [Intent Name]\n**Trigger:** (menu selection / keywords / both)  \n**Data required:** \n\n```\nStep 1 — Acknowledge:\n\nStep 2 — Data collect:\n\nStep 3 — Lookup/check:\n\nStep 4A — Resolve (success):\n\nStep 4B — Cannot resolve:\n\nStep 5 — Escalation:\n\nStep 6 — Confirmation:\n```\n\n### Flow 2: [Intent Name]\n**Trigger:**  \n**Data required:** \n\n```\n[Repeat structure]\n```\n\n## Escalation Rules\n| Trigger | Rule | Action |\n|---|---|---|\n| Frustration keywords | \"ridiculous\", \"furious\", \"unacceptable\", \"worst\" | Immediate human handoff |\n| 2 failed resolution attempts | Any intent | Escalate with transcript |\n| Explicit request | \"human\", \"agent\", \"real person\" | Immediate handoff |\n| High-value customer | LTV > $[threshold] | Priority queue |\n\n## Response Time Commitments (communicated to customer)\n- Business hours escalation: respond within ___ hours\n- After-hours escalation: respond by ___ next business day\n- Acknowledgement message: sent immediately upon escalation\n\n## Measurement Plan\n| KPI | Baseline | 30-day target | 90-day target |\n|---|---|---|---|\n| Containment rate | | >50% | >65% |\n| Escalation rate | | <40% | <30% |\n| CSAT (bot) | | >3.5/5 | >3.8/5 |\n| Drop-off rate | | <20% | <15% |\n\n## Unhandled Intent Review Schedule\n- Weekly review: Mondays, 30 minutes\n- New flow threshold: intent appears 10+ times in a week → add to build queue\n\nFile v1.1.0:quality-checklist.md\n\n# Chatbot Designer Quality Checklist\n\n## Pre-Build\n- [ ] 90 days of support ticket data pulled and categorized by intent\n- [ ] Top 5 intents by volume identified (not assumed)\n- [ ] Current first-response time baseline recorded\n- [ ] CSAT baseline recorded (to compare post-launch)\n- [ ] OMS/platform integration confirmed (can bot look up live order data?)\n\n## Flow Design\n- [ ] Main menu has no more than 6 options\n- [ ] \"Talk to a person\" option always visible\n- [ ] Every flow has a clear resolution path (bot resolves OR creates ticket — never dead-ends)\n- [ ] Every flow has an escalation exit after 2 failed resolution attempts\n- [ ] Escalation pre-fills ticket with full conversation transcript\n- [ ] Response time communicated to customer at escalation (specific time, not \"soon\")\n- [ ] No flow requires more data than necessary (minimum data collection per flow)\n\n## Specific Flows\n- [ ] Order status flow connected to live OMS data\n- [ ] Return flow checks eligibility against order date + policy window\n- [ ] Return flow generates or emails prepaid label automatically\n- [ ] Cancellation flow checks fulfillment status before confirming cancellation\n- [ ] Wrong item flow initiates replacement + return at no cost\n- [ ] Pre-purchase compatibility questions route to FAQ lookup or human\n\n## Escalation Rules\n- [ ] Frustration keywords trigger immediate human escalation\n- [ ] High-value customer tag routes to priority queue\n- [ ] Explicit \"human/agent\" request triggers immediate handoff\n- [ ] After-hours escalation gives specific response time (not vague)\n- [ ] Safety/legal keywords route to senior/priority queue\n\n## Copy and Tone\n- [ ] Bot introduces itself as a bot (not pretending to be human)\n- [ ] Contractions used (\"I'll\", \"you're\", \"let's\")\n- [ ] Specific numbers used (not \"soon\" or \"quickly\")\n- [ ] Frustration acknowledged before jumping to solution\n- [ ] No robotic completion phrases (\"Is there anything else I can help you with today?\")\n- [ ] Error messages always include next step (never dead-end)\n\n## Technical\n- [ ] Chatbot tested end-to-end for every flow\n- [ ] Escalation tested: frustration keywords trigger handoff correctly\n- [ ] Mobile display tested (readable, tappable, no horizontal scroll)\n- [ ] OMS integration confirmed with live order lookup test\n- [ ] CSAT survey configured (20–30% random sample, not every interaction)\n- [ ] Unhandled intent logging active\n\n## Measurement Setup\n- [ ] Containment rate tracking configured\n- [ ] Escalation rate tracking configured\n- [ ] CSAT for bot interactions tracking configured\n- [ ] Drop-off rate tracking configured\n- [ ] Weekly unhandled intent review scheduled (calendar block)\n- [ ] 30-day and 90-day targets documented\n\nFile v1.1.0:skill-card.md\n\n## Description:\n\nDesign customer service chatbot conversation flows for ecommerce - order status, returns, product recommendations, and escalation rules - that reduce ticket volume while maintaining satisfaction scores.\n\nThis skill is ready for commercial/non-commercial use.\n\n## Publisher:\n\n[leooooooow](https://clawhub.ai/user/leooooooow)\n\n### License/Terms of Use:\n\nMIT-0\n\n## Use Case:\n\nExternal ecommerce support teams and chatbot builders use this skill to design customer service chatbot flows, intent maps, escalation rules, response copy, and measurement plans for order status, returns, product questions, billing issues, and human handoff.\n\n### Deployment Geography for Use:\n\nGlobal\n\n## Known Risks and Mitigations:\n\nRisk: The skill recommends sharing full conversation transcripts during escalation, which may expose unnecessary customer data.\n\nMitigation: Use minimized and redacted handoff summaries unless the privacy policy, consent flow, retention rules, and support access controls explicitly allow broader transcript sharing.\n\nRisk: Automated cancellations, refunds, replacements, return labels, and claims can affect customer accounts or orders if triggered without confirmation.\n\nMitigation: Require explicit customer confirmation before any account, order, refund, replacement, return-label, or claim action is executed.\n\nRisk: Broad single-word intent triggers can route customers into order-data flows too easily.\n\nMitigation: Use contextual intent matching and confirmation prompts before collecting order data or initiating account-specific workflows.\n\n## Reference(s):\n\n- [ClawHub skill page](https://clawhub.ai/leooooooow/skills/chatbot-designer)\n- [Output Template](artifact/output-template.md)\n- [Intent Library](artifact/intent-library.md)\n- [Escalation Playbook](artifact/escalation-playbook.md)\n- [Quality Checklist](artifact/quality-checklist.md)\n\n## Skill Output:\n\n**Output Type(s):** [text, markdown, guidance, configuration]\n\n**Output Format:** [Markdown guidance with flow templates, decision tables, checklist items, and conversation copy examples]\n\n**Output Parameters:** [1D]\n\n**Other Properties Related to Output:** [Produces chatbot flow design documents, intent mappings, escalation protocols, quality checklists, and response templates for ecommerce support workflows.]\n\n## Skill Version(s):\n\n1.1.0 (source: server release evidence)\n\n## Ethical Considerations:\n\nUsers should evaluate whether this skill is appropriate for their environment, review any generated or modified files before relying on them, and apply their organization's safety, security, and compliance requirements before deployment.\n\nArchive v1.0.0: 3 files, 4352 bytes\n\nFiles: skill-card.md (2230b), SKILL.md (6303b), _meta.json (135b)\n\nFile v1.0.0:SKILL.md\n\n---\nname: chatbot-designer\ndescription: Design customer service chatbot conversation flows for ecommerce including order status inquiries, return requests, product recommendations, and escalation rules that reduce ticket volume while maintaining satisfaction scores.\n---\n\n# Chatbot Designer\n\nEcommerce customer service teams drown in repetitive tickets — \"where's my order,\" \"how do I return this,\" \"does this come in blue\" — while high-value conversations that need human nuance get buried in the queue. A well-designed chatbot handles the predictable inquiries instantly and routes the complex ones to agents with full context. But most chatbot implementations fail because the conversation flows are designed by engineers guessing at customer intent rather than being mapped from actual support patterns. This skill designs complete chatbot conversation architectures grounded in ecommerce-specific support workflows, so you launch a bot that genuinely deflects tickets instead of frustrating customers into demanding a human.\n\n## Use when\n\n- You are setting up a customer service chatbot for your Shopify, Amazon, or TikTok Shop store and need conversation flow diagrams before building in your chatbot platform\n- A CX manager says \"we need to reduce our ticket volume by 40% without hurting our CSAT score — design me a chatbot that can handle the top inquiry types\"\n- You are migrating from a basic FAQ bot to a multi-turn conversational chatbot and need structured dialogue trees for order tracking, returns, exchanges, product questions, and escalation paths\n- Your current chatbot has a high abandonment rate or excessive escalation rate and you need to redesign the flows to actually resolve inquiries\n\n## What this skill does\n\nThis skill takes your store's support context — top inquiry categories, product types, return policy, shipping carriers, and any platform-specific constraints — and generates a complete chatbot conversation architecture. It designs multi-turn dialogue flows for each major inquiry type, mapping out the decision tree from initial customer message through resolution or escalation. Each flow includes intent detection triggers (the phrases and keywords that route customers into the right flow), clarifying question sequences, API integration points where the bot needs to pull live data (order status, tracking numbers, inventory availability), response templates with personalization placeholders, and explicit escalation criteria that define when and how to hand off to a human agent. The skill also designs fallback handling for unrecognized intents, satisfaction measurement touchpoints, and a conversation analytics framework so you can measure and iterate on bot performance after launch.\n\n## Inputs required\n\n- **Top support inquiry categories** (required): The most common types of customer inquiries your team handles, ranked by volume. Example: \"1. Order status/tracking (35%), 2. Returns and exchanges (25%), 3. Product sizing questions (15%), 4. Shipping time estimates (10%), 5. Discount code issues (8%), 6. Other (7%)\"\n- **Store policies summary** (required): Your return policy, shipping policy, and any warranty or guarantee terms the bot needs to communicate accurately. Example: \"30-day returns, free return shipping on defective items, customer pays return shipping on preference returns, 3-5 business day standard shipping\"\n- **Product catalog context** (required): A brief description of what you sell and any product-specific FAQ patterns. Example: \"Women's athletic wear, sizes XS-3XL. Common questions: fabric composition, size chart accuracy, sports bra support level, washing instructions\"\n- **Chatbot platform** (optional): The tool you plan to build in (Tidio, Gorgias, Zendesk, Intercom, custom). If specified, the skill tailors integration recommendations and flow formatting to that platform's capabilities.\n- **Current escalation rate or CSAT score** (optional): If you have baseline metrics from an existing bot or live chat, providing them helps the skill set realistic improvement targets and prioritize which flows to optimize first.\n\n## Output format\n\nThe output is structured as a Chatbot Architecture Document with five major sections. The first section is an Intent Map listing all supported customer intents with their trigger phrases, confidence thresholds, and routing priorities. The second section contains Conversation Flow Diagrams for each major intent — presented as structured decision trees showing each bot message, expected customer responses, branching logic, API call points, and terminal states (resolved, escalated, or abandoned). Each node in the flow includes the exact message template with personalization variables marked in brackets. The third section is an Escalation Rules Matrix defining the conditions that trigger human handoff, the context data passed to the agent, priority levels, and SLA targets for each escalation type. The fourth section covers Fallback and Edge Case Handling including unrecognized intent responses, repeated failure loops, profanity or frustration detection, and after-hours behavior. The fifth section is an Analytics and Iteration Framework specifying which KPIs to track (resolution rate, escalation rate, CSAT per flow, average turns to resolution) and a suggested A/B testing roadmap for optimizing underperforming flows. The document is designed to serve as a complete specification that a developer or chatbot platform administrator can implement directly.\n\n## Scope\n\n- Designed for: ecommerce operators, customer experience managers, support team leads, and chatbot developers\n- Platform context: Shopify, Amazon Seller Central, TikTok Shop, WooCommerce, and platform-agnostic DTC stores\n- Language: English\n\n## Limitations\n\n- Does not build or deploy the actual chatbot — this skill produces the conversation architecture and flow specifications, not executable bot code\n- Does not integrate with your live order management or CRM system to pull real customer data during the design phase; API integration points are specified but must be connected during implementation\n- Escalation rate reduction estimates are based on industry benchmarks for ecommerce chatbots and may vary based on your specific customer base complexity and product category\n\nFile v1.0.0:_meta.json\n\n{\n  \"ownerId\": \"kn70fv0ehp50emedet9tx3fekd82pw3b\",\n  \"slug\": \"chatbot-designer\",\n  \"version\": \"1.0.0\",\n  \"publishedAt\": 1776388133546\n}\n\nFile v1.0.0:skill-card.md\n\n## Description: <br>\nDesign customer service chatbot conversation flows for ecommerce including order status inquiries, return requests, product recommendations, and escalation rules that reduce ticket volume while maintaining satisfaction scores. <br>\n\nThis skill is ready for commercial/non-commercial use. <br>\n\n## Publisher: <br>\n[leooooooow](https://clawhub.ai/user/leooooooow) <br>\n\n### License/Terms of Use: <br>\nMIT-0 <br>\n\n\n## Use Case: <br>\nEcommerce operators, customer experience managers, support team leads, and chatbot developers use this skill to design chatbot intent maps, conversation flows, escalation rules, fallback handling, and analytics plans before implementing a customer service bot. <br>\n\n### Deployment Geography for Use: <br>\nGlobal <br>\n\n## Known Risks and Mitigations: <br>\nRisk: Real customer PII, live order exports, API tokens, or private credentials could be included in prompts while preparing chatbot flows. <br>\nMitigation: Use sanitized support patterns and policy summaries, and keep credentials and live customer records out of prompts. <br>\nRisk: Generated chatbot flows may be inaccurate or unsuitable for a live customer-facing support system if implemented without review. <br>\nMitigation: Review the generated architecture, policies, escalation criteria, and message templates before implementing them in production. <br>\n\n\n## Reference(s): <br>\n\n\n## Skill Output: <br>\n**Output Type(s):** [Markdown, Configuration instructions, Guidance] <br>\n**Output Format:** [Structured Chatbot Architecture Document with intent maps, conversation flow diagrams, escalation matrices, fallback handling, and analytics recommendations] <br>\n**Output Parameters:** [1D] <br>\n**Other Properties Related to Output:** [Produces design specifications and implementation guidance; it does not build, deploy, or connect the chatbot to live systems.] <br>\n\n## Skill Version(s): <br>\n1.0.0 (source: server release metadata) <br>\n\n## Ethical Considerations: <br>\nUsers should evaluate whether this skill is appropriate for their environment, review any generated or modified files before relying on them, and apply their organization's safety, security, and compliance requirements before deployment. <br>","readmeExcerpt":"Skill: Chatbot Designer Owner: leooooooow Summary: Design customer service chatbot conversation flows for ecommerce — order status, returns, product recommendations, and escalation rules — that reduce ticket... Tags: latest:1.1.0 Version history: v1.1.0 | 2026-06-05T14:44:21.110Z | user **Changelog – Version 1.0.1** - Added four documentation files: escalation-playbook.md, intent-library.md, output-template.md, and q","codeSnippets":[],"executableExamples":[{"language":"text","snippet":"Hi [name]! How can I help today?\n  → 📦 Track my order\n  → 🔄 Return or exchange\n  → ❓ Product question\n  → 💳 Billing or payment\n  → 🙋 Talk to a person"},{"language":"text","snippet":"1. TRIGGER: Intent detected (menu selection or keyword match)\n2. ACKNOWLEDGE: \"Let me pull that up for you.\"\n3. DATA COLLECT: What does the bot need? (order number, email, product name)\n4. LOOKUP / CHECK: Connect to data source or apply policy rules\n5. RESOLUTION BRANCH A: Can resolve → confirm resolution, offer next step\n6. RESOLUTION BRANCH B: Cannot resolve → explain why + escalate gracefully\n7. ESCALATION: Create ticket with full context pre-filled; set expectation (\"Team will reply in 2–4 hours\")\n8. CONFIRMATION: Always confirm what happened before ending conversation"},{"language":"text","snippet":"Bot: \"To track your order, I'll need your order number or the email you used to order.\"\nCustomer: [provides order #12345]\nBot: [looks up OMS] → \"Order #12345 was shipped on June 3rd via FedEx.\n     Estimated delivery: June 7th.\n     Tracking: [link]\n     Is there anything else you'd like to know about this order?\"\n     → Yes → return to order menu\n     → No → \"Thanks! If your package doesn't arrive by June 8th, come back and I'll help you file a claim.\""},{"language":"text","snippet":"Customer: \"I want to return my order\"\nBot: \"I can help with that. What's your order number?\"\n→ [Order number lookup]\n→ Check: Is order within return window? (e.g., 30 days from delivery)\n   → YES: \"Great — this order is eligible. What's the reason for your return?\"\n           → [Reason menu: wrong size / doesn't meet expectations / damaged / wrong item]\n           → [Generate return label or RMA number]\n           → \"Your prepaid return label has been sent to [email]. Please drop off within 7 days.\"\n   → NO: \"This order was delivered on [date] — our 30-day window closed on [date].\n           I'm not able to process this automatically, but I'll connect you with our team\n           who can review exceptions.\"\n           → [Create ticket: late return request, order details pre-filled]"},{"language":"text","snippet":"\"What device are you trying to use this with?\"\n→ [Device selection menu: iPhone 15 / iPhone 14 / Samsung S24 / Other]\n→ \"The [product] is fully compatible with [device]. It includes [cable type] in the box.\n   Want to add it to your cart?\""},{"language":"text","snippet":"[BOT HANDOFF] Customer: [Name] | Order: #[number]\nIssue: [1-sentence summary]\nBot actions taken: [what was attempted]\nData collected: [relevant fields]\nCustomer sentiment: [frustrated/neutral/satisfied]\nChat transcript: [attached]"}],"parameters":null,"dependencies":[],"permissions":[],"extractedFiles":[{"path":"SKILL.md","content":"---\nname: Chatbot Designer\ndescription: Design customer service chatbot conversation flows for ecommerce — order status, returns, product recommendations, and escalation rules — that reduce ticket volume while maintaining satisfaction scores.\n---\n\n# Chatbot Designer\n\nDesign customer service chatbot conversation flows for ecommerce including order status inquiries, return requests, product recommendations, and escalation rules that reduce ticket volume while maintaining satisfaction scores. Most chatbot failures come not from the technology but from poorly designed flows — dead ends, missing escalation paths, or responses that feel robotic. This skill produces conversation architecture, intent mapping, and response logic that resolves common queries automatically while seamlessly handing off complex issues to human agents.\n\n## Quick Reference\n\n| Decision | Strong | Acceptable | Weak |\n|---|---|---|---|\n| Intent coverage | Map top 10 intents from ticket data before writing any flows | Map intents from assumption | Build flows for every possible scenario before launching |\n| Escalation design | Graceful escalation after 2 failed responses + any customer frustration signal | Escalation only when explicitly requested | No escalation path — bot loops or dead-ends |\n| Tone | Warm, direct, brand-consistent; acknowledges frustration | Neutral corporate tone | Overly casual/emoji-heavy OR stiff formal robospeak |\n| Resolution path | Bot resolves or creates ticket — never leaves customer waiting without next step | Resolves common issues, drops others | Gives information without action (\"call this number\") |\n| Order integration | Connected to OMS so bot can pull live order status | Static FAQs about shipping timelines | Cannot access order data at all |\n| Return flow | Initiates return label automatically; confirms eligibility in real time | Explains return policy; directs to email | Can't process returns; sends customer elsewhere |\n| Measurement | CSAT on bot interactions + containment rate + escalation rate tracked weekly | Track containment rate only | No metrics |\n\n## Solves\n\n- Support team overwhelmed with high-volume, repetitive queries (order status, return requests, shipping ETAs)\n- Long first-response times hurting customer satisfaction and review scores\n- After-hours support gaps — customers getting no response outside business hours\n- Support costs scaling linearly with order volume instead of leveling off\n- Inconsistent answers from human agents on standard policy questions\n- High ticket volume from customers who can't self-serve on the website\n- Returns and exchanges handled manually when they could be automated end-to-end\n\n## Workflow\n\n### Step 1 — Map Your Top 10 Support Intents\n\nPull 90 days of support ticket data and categorize by intent. Most ecommerce stores find 80% of volume concentrated in 5–7 intents.\n\n**Typical ecommerce intent distribution:**\n\n| Intent | Average % of tickets | Automatable? |\n|---|---|---|\n| Order status / tracking |"},{"path":"_meta.json","content":"{\n  \"ownerId\": \"kn70fv0ehp50emedet9tx3fekd82pw3b\",\n  \"slug\": \"chatbot-designer\",\n  \"version\": \"1.1.0\",\n  \"publishedAt\": 1780670661110\n}"},{"path":"escalation-playbook.md","content":"# Escalation Playbook\n\n## Escalation Philosophy\n\nAn escalation is not a failure — it is the chatbot doing its job. The goal is to:\n1. Catch the moment when a human is needed\n2. Transfer context so the customer doesn't repeat themselves\n3. Set an accurate expectation for response time\n4. Leave the customer feeling helped, not abandoned\n\nA chatbot with excellent escalation design achieves higher CSAT than one that tries to resolve everything and gets stuck.\n\n## Escalation Trigger Taxonomy\n\n### Tier 1 — Immediate Escalation (within seconds)\n\n| Trigger | Detection method | Queue |\n|---|---|---|\n| Physical safety mention | Keywords: injury, hurt, damaged, burned, allergic reaction | Emergency / Senior |\n| Legal or fraud mention | Keywords: fraud, dispute, chargeback, lawyer, report | Legal / Senior |\n| High-value customer | CRM tag: LTV > $[threshold] | Priority |\n| Explicit human request | Keywords: human, agent, person, real | General |\n| Severe frustration language | Keywords: furious, unacceptable, disgusting, lawsuit | Priority |\n\n### Tier 2 — Escalation After 2 Failed Attempts\n\n| Trigger | Condition | Queue |\n|---|---|---|\n| Unresolvable intent | Bot attempted twice, no resolution | General |\n| Missing order data | Cannot find order after 2 lookups | General |\n| Out-of-policy request | Outside return window, discontinued product | General |\n| Complex multi-issue | Customer raises 3+ different issues in one session | General |\n\n### Tier 3 — Proactive Escalation (Batched)\n\n| Trigger | Timing | Action |\n|---|---|---|\n| Unhandled intent | Weekly review | Add to flow build queue |\n| Low CSAT bot rating | <3/5 rating given | Review transcript; identify gap |\n| Long session with no resolution | Session > 10 min, no confirmation | Flag for quality review |\n\n## Escalation Handoff Protocol\n\n### Step 1 — Signal the transition warmly\nNever abruptly transfer. Use a bridging line:\n> \"Let me get one of our team members to help you with this.\"\n> \"This is something I want to make sure our specialists handle for you.\"\n> \"I'm connecting you now — you're in good hands.\"\n\n### Step 2 — Summarize the conversation\nPre-fill the ticket with:\n- Customer name and contact\n- Order number(s) involved\n- Issue summary in 2–3 sentences\n- What the bot already tried\n- Customer sentiment (frustrated / neutral / positive)\n- Any data already collected (return reason, device type, etc.)\n\n**Ticket pre-fill template:**\n```\n[BOT HANDOFF] Customer: [Name] | Order: #[number]\nIssue: [1-sentence summary]\nBot actions taken: [what was attempted]\nData collected: [relevant fields]\nCustomer sentiment: [frustrated/neutral/satisfied]\nChat transcript: [attached]\n```\n\n### Step 3 — Set response time expectation\nNever use vague language. Be specific:\n- ✅ \"Our team will reply within 2 business hours\"\n- ✅ \"You'll hear back by 9am tomorrow\"\n- ❌ \"Someone will be in touch soon\"\n- ❌ \"We'll reply as quickly as possible\"\n\n### Step 4 — Confirm and close the bot interaction\n> \"I've sent everything to our team an"},{"path":"intent-library.md","content":"# Intent Library — Ecommerce Chatbot\n\n## Order Status Intent\n\n**Trigger keywords:** \"where is my order\", \"track\", \"tracking\", \"shipping status\", \"hasn't arrived\", \"order status\", \"when will it arrive\", \"delivery\"\n\n**Required data:** Order number OR email address\n\n**Success response template:**\n> \"I found your order! Order #[number] shipped on [date] via [carrier]. Your tracking number is [tracking] — click here to see live updates: [link]. Estimated delivery: [date]. Is there anything else I can help with?\"\n\n**Delayed order response:**\n> \"Your order #[number] is currently showing a delay with [carrier]. The new estimated delivery is [date]. I've flagged this on your account. If it doesn't arrive by [date+1], come back and I can file a claim on your behalf.\"\n\n**Not found response:**\n> \"I couldn't find an order with that information. Can you double-check the order number? It's usually in your confirmation email and starts with #. If you still can't find it, let me connect you with our team.\"\n\n---\n\n## Return Request Intent\n\n**Trigger keywords:** \"return\", \"refund\", \"exchange\", \"send back\", \"doesn't fit\", \"changed my mind\", \"not what I expected\"\n\n**Required data:** Order number, reason for return\n\n**Eligible return response:**\n> \"I can process this return for you. Your order is within our [30-day] return window. I'll send a prepaid return label to [email] — please allow 5–10 minutes for it to arrive. Once we receive your return, your refund will be processed within [X] business days.\"\n\n**Ineligible return response (outside window):**\n> \"This order was delivered on [date] — our [30-day] return window closed on [date]. I'm not able to process this automatically, but I'll connect you with our team who can review your situation.\"\n\n**Exchange response:**\n> \"For an exchange, I'll need to process a return and place a new order. Would you like me to start the return process? You can then place a new order with the correct [size/color/item], and I can apply a discount code for the inconvenience.\"\n\n---\n\n## Shipping Delay Intent\n\n**Trigger keywords:** \"late\", \"delayed\", \"hasn't shipped\", \"still processing\", \"stuck\", \"where\", \"overdue\"\n\n**Response template:**\n> \"I'm sorry to hear your order is running late. Let me check on this. [Look up order status] Your order is currently [status]. [If in transit:] It looks like there's been a carrier delay — the new estimated delivery is [date]. [If not yet shipped:] It's still being prepared for shipment. I'll flag this as high priority for our fulfillment team.\"\n\n---\n\n## Wrong Item Received Intent\n\n**Trigger keywords:** \"wrong item\", \"wrong product\", \"not what I ordered\", \"incorrect\", \"different item\"\n\n**Response template:**\n> \"I'm really sorry about that — receiving the wrong item is frustrating. I'd like to make this right. Can you share: (1) your order number, and (2) what you received vs. what you ordered? I'll initiate a replacement shipment and arrange return of the incorrect item at no cost to you.\"\n\n---\n\n## Product"},{"path":"output-template.md","content":"# Chatbot Flow Design Document\n\n## Project Overview\n- **Brand:** \n- **Platform:** Shopify / WooCommerce / Custom\n- **Chatbot tool:** Gorgias / Tidio / Intercom / Zendesk / Custom\n- **OMS integration:** Yes / No / Planned\n- **Launch date target:** \n\n## Intent Priority Matrix\n| Rank | Intent | % of ticket volume | Automation potential | Build priority |\n|---|---|---|---|---|\n| 1 | | % | High/Medium/Low | Launch / Phase 2 |\n| 2 | | % | | |\n| 3 | | % | | |\n| 4 | | % | | |\n| 5 | | % | | |\n\n## Main Menu Design\n```\nWelcome message: \n\nMenu options:\n1. \n2. \n3. \n4. \n5. Talk to a person\n```\n\n## Flow Designs\n\n### Flow 1: [Intent Name]\n**Trigger:** (menu selection / keywords / both)  \n**Data required:** \n\n```\nStep 1 — Acknowledge:\n\nStep 2 — Data collect:\n\nStep 3 — Lookup/check:\n\nStep 4A — Resolve (success):\n\nStep 4B — Cannot resolve:\n\nStep 5 — Escalation:\n\nStep 6 — Confirmation:\n```\n\n### Flow 2: [Intent Name]\n**Trigger:**  \n**Data required:** \n\n```\n[Repeat structure]\n```\n\n## Escalation Rules\n| Trigger | Rule | Action |\n|---|---|---|\n| Frustration keywords | \"ridiculous\", \"furious\", \"unacceptable\", \"worst\" | Immediate human handoff |\n| 2 failed resolution attempts | Any intent | Escalate with transcript |\n| Explicit request | \"human\", \"agent\", \"real person\" | Immediate handoff |\n| High-value customer | LTV > $[threshold] | Priority queue |\n\n## Response Time Commitments (communicated to customer)\n- Business hours escalation: respond within ___ hours\n- After-hours escalation: respond by ___ next business day\n- Acknowledgement message: sent immediately upon escalation\n\n## Measurement Plan\n| KPI | Baseline | 30-day target | 90-day target |\n|---|---|---|---|\n| Containment rate | | >50% | >65% |\n| Escalation rate | | <40% | <30% |\n| CSAT (bot) | | >3.5/5 | >3.8/5 |\n| Drop-off rate | | <20% | <15% |\n\n## Unhandled Intent Review Schedule\n- Weekly review: Mondays, 30 minutes\n- New flow threshold: intent appears 10+ times in a week → add to build queue"}],"languages":[],"docsSourceLabel":"CLAWHUB","editorialOverview":"Design customer service chatbot conversation flows for ecommerce — order status, returns, product recommendations, and escalation rules — that reduce ticket... Skill: Chatbot Designer Owner: leooooooow Summary: Design customer service chatbot conversation flows for ecommerce — order status, returns, product recommendations, and escalation rules — that reduce ticket... Tags: latest:1.1.0 Version history: v1.1.0 | 2026-06-05T14:44:21.110Z | user **Changelog – Version 1.0.1** - Added four documentation files: escalation-playbook.md, intent-library.md, output-template.md, and q","editorialQuality":{"score":100,"threshold":65,"status":"ready","wordCount":1811,"uniquenessScore":48,"reasons":[]}},"media":{"evidence":{"source":"no-media","verified":false,"confidence":"low","updatedAt":"2026-10-11T15:35:57.838Z","emptyReason":"No screenshots, media assets, or demo links are available."},"primaryImageUrl":null,"mediaAssetCount":0,"assets":[],"demoUrl":null},"ownerResources":{"evidence":{"source":"unclaimed","verified":false,"confidence":"low","updatedAt":"2026-10-11T15:35:57.838Z","emptyReason":"This page has not been claimed by the agent owner."},"hasCustomPage":false,"customPageUpdatedAt":null,"customLinks":[],"structuredLinks":{"docsUrl":null,"demoUrl":null,"supportUrl":null,"pricingUrl":null,"statusUrl":null},"customPage":null},"relatedAgents":{"evidence":{"source":"protocol-neighbors","verified":false,"confidence":"medium","updatedAt":"2026-10-11T17:43:01.782Z","emptyReason":null},"items":[{"id":"8ebccd8e-3863-4187-8355-c3f14e1f9edf","entityType":"agent","canonicalPath":"/agent/iofficeai-aionui","slug":"iofficeai-aionui","name":"AionUi","description":"Free, local, open-source 24/7 Cowork app and OpenClaw for Gemini CLI, Claude Code, Codex, OpenCode, Qwen Code, Goose CLI, Auggie, and more | 🌟 Star if you like it!","url":"https://github.com/iOfficeAI/AionUi","homepage":"https://www.aionui.com","source":"GITHUB_REPOS","protocols":["MCP","OPENCLAW"],"capabilities":[],"safetyScore":100,"overallRank":70,"updatedAt":"2026-10-09T19:11:12.944Z","createdAt":"2026-02-25T03:38:16.584Z","downloads":null},{"id":"b917f68a-ebff-438e-84f8-3f4b2494c0bc","entityType":"agent","canonicalPath":"/agent/activepieces-activepieces","slug":"activepieces-activepieces","name":"activepieces","description":"AI Agents & MCPs & AI Workflow Automation • (~400 MCP servers for AI agents) • AI Automation / AI Agent with MCPs • AI Workflows & AI Agents • MCPs for AI Agents","url":"https://github.com/activepieces/activepieces","homepage":"https://www.activepieces.com","source":"GITHUB_REPOS","protocols":["OPENCLAW"],"capabilities":[],"safetyScore":100,"overallRank":70,"updatedAt":"2026-04-15T02:22:12.426Z","createdAt":"2026-02-25T03:38:12.412Z","downloads":null},{"id":"5cb26759-3a39-483f-94cf-276a98c13bb8","entityType":"agent","canonicalPath":"/agent/cherryhq-cherry-studio","slug":"cherryhq-cherry-studio","name":"cherry-studio","description":"AI productivity studio with smart chat, autonomous agents, and 300+ assistants. Unified access to frontier LLMs","url":"https://github.com/CherryHQ/cherry-studio","homepage":"https://cherry-ai.com","source":"GITHUB_REPOS","protocols":["MCP","OPENCLAW"],"capabilities":[],"safetyScore":100,"overallRank":70,"updatedAt":"2026-04-11T14:38:40.986Z","createdAt":"2026-02-25T03:38:19.379Z","downloads":null},{"id":"6f6582d0-5d76-4f0f-b81d-86520247950b","entityType":"agent","canonicalPath":"/agent/copilotkit-copilotkit","slug":"copilotkit-copilotkit","name":"CopilotKit","description":"The Frontend for Agents & Generative UI. React + Angular","url":"https://github.com/CopilotKit/CopilotKit","homepage":"https://docs.copilotkit.ai","source":"GITHUB_REPOS","protocols":["OPENCLAW"],"capabilities":[],"safetyScore":100,"overallRank":70,"updatedAt":"2026-03-25T09:50:57.846Z","createdAt":"2026-02-25T03:39:14.617Z","downloads":null}],"links":{"hub":"/agent","source":"/agent/source/clawhub","protocols":[{"label":"OpenClaw","href":"/agent/protocol/openclew"}]}}}