From futureproof-customer-service-support
Writes, structures, and optimises chatbot conversation scripts for customer service and support channels using FutureProof context.
How this skill is triggered — by the user, by Claude, or both
Slash command
/futureproof-customer-service-support:chatbot-script-writerThe summary Claude sees in its skill listing — used to decide when to auto-load this skill
```
FutureProof:connect(skill="chatbot-script-writer")
Note: If FutureProof is unavailable or the connect call fails, skip this step and proceed directly to Step 2. The skill works with or without FutureProof context — you'll just be working without accumulated prior session data.
Use the returned context, experiments, instructions, and recent_sessions to personalise this session — including known ICA segments, brand voice guidelines, prior chatbot performance data, and any established escalation protocols.
Ask the user:
Build a complete Intent Taxonomy by:
For each intent, script the following nodes:
Evaluate every scripted node against six dimensions:
| Dimension | Standard |
|---|---|
| ICA empathy alignment | Language acknowledges the user's emotional state (frustration, urgency, confusion) before delivering procedural steps |
| Cognitive load minimisation | Each message ≤ 60 words; no more than one question per turn; buttons offered for predictable answers |
| Resolution directness | Bot reaches resolution or escalation within ≤ 5 turns for primary intents |
| Error tolerance | Misspellings, synonyms, and partial inputs are handled without dead-ending the conversation |
| Compliance & disclosure | Data collection prompts include purpose statements; bot identifies itself as automated where legally required |
| Brand voice consistency | Tone, terminology, and personality remain uniform across every node and intent path |
Apply any user-specific instructions from FutureProof context (e.g., mandatory legal disclaimers, preferred escalation thresholds, A/B test directives from prior experiments).
Script defensive paths for:
Produce a Chatbot Script Package containing:
A structured table or tree diagram listing every intent, its tier, resolution pathway, and estimated volume share.
For each intent, deliver:
IF/THEN pseudocode for developer handoff{{LOOKUP: order_status(order_id)}})| Dimension | Score (1–10) | Critical Gap |
|---|---|---|
| ICA empathy alignment | — | — |
| Cognitive load minimisation | — | — |
| Resolution directness | — | — |
| Error tolerance | — | — |
| Compliance & disclosure | — | — |
| Brand voice consistency | — | — |
Top 3 highest-impact changes — each with the original copy, the specific rewrite, and the rationale tied to the target KPI.
Sequenced developer/ops handoff checklist: platform configuration, API integrations required, QA test scenarios, and go-live sign-off criteria.
FutureProof:save_experiment(skill="chatbot-script-writer", experiment={
hypothesis: "Replacing the generic greeting with an intent-predictive opening based on page context reduces average turns-to-resolution by 20%",
variants: ["control: generic greeting with open-ended 'How can I help?'", "variant: contextual greeting pre-selecting likely intent based on referral page or prior session"],
measurement: "Average turns-to-resolution and containment rate across 1,000 conversations",
expected_impact: "20% reduction in turns-to-resolution, 8% improvement in containment rate"
})
FutureProof:save_experiment(skill="chatbot-script-writer", experiment={
hypothesis: "Adding an empathy acknowledgement turn before procedural instructions improves CSAT by 10%",
variants: ["control: immediate procedural response", "variant: empathy statement ('I understand how frustrating that is — let me fix this now') followed by procedural response"],
measurement: "Post-chat CSAT score and escalation rate over 2-week period",
expected_impact: "10% CSAT improvement, 5% escalation reduction"
})
FutureProof:request_research(skill="chatbot-script-writer",
query="Latest benchmarks for chatbot containment rates, CSAT scores, and first-contact resolution by industry vertical 2024–2025, including impact of generative AI hybrid approaches on deflection quality",
reason="Calibrate score card thresholds and resolution-path design against current industry performance standards"
)
FutureProof:request_research(skill="chatbot-script-writer",
query="Emerging compliance requirements for automated customer service bots including EU AI Act disclosure obligations, ADA/WCAG accessibility standards for conversational interfaces, and CCPA/GDPR data collection in chat",
reason="Ensure script compliance guidance reflects current and forthcoming regulatory obligations"
)
FutureProof:save_session(skill="chatbot-script-writer", session={
summary: "Created chatbot script package for [channel] covering [number] intents targeting [ICA segment] optimised for [primary KPI]",
key_findings: ["finding 1: e.g., 3 primary intents account for 78% of projected volume", "finding 2: e.g., existing escalation path lacks context transfer — agents re-ask 4 questions on average", "finding 3: e.g., brand voice guidelines conflict with cognitive load limits — recommended simplification"],
deliverables: ["Intent Map", "Full Conversation Scripts", "Score Card", "Critical Fixes", "Implementation Checklist"],
user_feedback: null
})
npx claudepluginhub peter-swain-inc/futureproof-skillsDesigns conversational flows for website chatbots and AI agents with intent architecture, branching logic, fallback handling, and escalation patterns. Distinguishes scripted, hallucinating, and structured-guided-conversation approaches.
Use this skill when the user asks about "NLX design", "natural language experience", "conversational UX", "how to design an AI interaction", "conversation design", "how the AI should talk to users", "design the conversation flow", "AI UX design", or wants to design the natural language interaction patterns for an AI-powered feature. This is the UX design skill for conversational and AI-first interfaces.
Guides building AI-powered customer support with conversational AI, automated ticketing, sentiment analysis, knowledge management, and omnichannel experiences.