UNPKG

major-ai-skills

Version:

Installable agentic skills / AI agent skills (SKILL.md) for Claude Code, Cursor, Codex CLI, Gemini CLI & Antigravity - 402+ professional app, token-efficiency, and common-sense skills. SEO/GEO ready.

98 lines (71 loc) 4.59 kB
--- name: turn-complaints-into-polite-replies description: "Draft customer complaint replies that acknowledge the issue and provide concrete next steps without unsupported promises." category: common-sense risk: safe source: self source_type: self date_added: "2026-08-26" tags: ["customer-support", "de-escalation", "conflict-resolution", "hear-framework", "empathy", "prompt-engineering"] tools: ["claude", "cursor", "gemini", "codex", "chatgpt"] --- # Turn Complaints into Helpful Replies (The HEAR Framework) (AI Skill) ## Overview When faced with an aggressive, angry customer email, human support agents often respond with emotional defensiveness (*"As our policy clearly states..."*) or robotic, dismissive template replies (*"We apologize for any inconvenience caused"*). Both reactions escalate customer anger and lead to negative public reviews or chargebacks. The **HEAR De-Escalation Protocol** uses AI to instantly neutralize emotional friction, validate the customer's frustration, and present a clear, concrete path to resolution. --- ## The 4-Part HEAR De-Escalation Pipeline ``` ┌─────────────────────────────────────────────────────────────┐ The HEAR De-Escalation Loop [ H: HEAR & VALIDATE ] ──► Acknowledge the exact impact [ E: EMPATHIZE ] ──► Validate emotion, no excuses [ A: ACTIONABLE REMEDY ]──► Concrete fix + compensation [ R: REASSURE & TRACK ] ──► Systemic fix to prevent repeat └─────────────────────────────────────────────────────────────┘ ``` --- ## Master Complaint De-Escalation Prompt Templates ### Pattern 1: The Enterprise HEAR Resolution Engine Use for high-stakes customer or client complaints: ```markdown Here is an angry email from a customer: [PASTE EMAIL]. Context: - What happened: [EXPLAIN THE INTERNAL REALITY / GLITCH] - What we can offer: [e.g. Full refund / Free month credit / Priority patch] Draft a response following the **HEAR Framework**: 1. **Hear**: Acknowledge specifically what went wrong (do not use generic language). 2. **Empathize**: Show sincere human empathy without making excuses or admitting legal liability. 3. **Action**: Clearly detail the immediate remedy being applied today. 4. **Reassure**: Explain the concrete step taken so this never happens again. Constraints: - Warm, professional, and accountable tone. - Under 150 words. ``` --- ## Real-World Case Study ### Scenario: Client Furious Over Late Deliverable #### Raw Customer Email > *"This is completely unacceptable. You promised our marketing landing page would be live by 9 AM this morning. Our ad campaign is running right now and burning money! I want an immediate refund and an explanation!"* #### AI HEAR Response Output ```markdown Subject: Update on your landing page & immediate account credit Hi Mark, You are completely right to be frustrated. Having ad traffic running without the landing page live is unacceptable, and I understand the urgency. Here is what we are doing immediately: 1. **Going Live Now**: Our team has pushed the final assets live - your landing page is active and converting as of 9:45 AM. 2. **Ad Spend Credit**: We have applied a $500 credit to your account to offset the ad spend during this morning's 45-minute delay. 3. **Prevention**: We have updated our launch deployment checklist to require DNS verification 24 hours prior to ad launch. Please verify the page on your end here: [LINK]. I am monitoring the server traffic closely this morning to ensure 100% stability. Sincerely, [Your Name] ``` --- ## The 3 Golden Rules of Complaint Handling - **Never say "As per our policy"**: Policy citations make customers feel dismissed. Lead with solutions. - **Own the impact, not just the intent**: Acknowledge the customer's lost time or money directly. - **Give them a specific next step**: Always close with a link or action to verify the resolution.