本文へ移動
cccskills
無料GitHub で公開

helpdesk-responder

Generic helpdesk first-response skill for creating KB-based response drafts. Use when handling support tickets, creating response templates, or building a structured helpdesk workflow. Supports error code detection, keyword matching, confidence scoring, multi-language templates, and escalation workflows. Customize by providing your own KB articles and configuration.

インストール方法を見る

含まれるファイル(3)

  • SKILL.md13.5 KB
  • assets/response_templates.md11.5 KB
  • references/kb_schema.json9.7 KB

SKILL.md(原文)

インストールする前に、エージェントに与えられる指示の中身を確認できます。

Helpdesk First Response Skill

A generic helpdesk first-response skill that creates KB-based response drafts for support tickets. Adaptable to any industry or product support context.

Trigger Conditions

Use this skill when:

  • Handling customer support tickets or inquiries
  • Creating response drafts based on knowledge base articles
  • Building a structured helpdesk workflow
  • Training support staff on response patterns

Skill Workflow

+---------------------------------------------------------------------+
|                    Phase 1: Inquiry Analysis                        |
|  - Extract ticket information                                       |
|  - Auto-detect patterns (error codes, device names, symptoms)       |
+---------------------------------------------------------------------+
                              |
              +---------------+---------------+
              v               v               v
    +-------------+   +-------------+   +-------------+
    | Error Code  |   | Device/     |   | Keyword     |
    | Detection   |   | Product     |   | Detection   |
    +-------------+   +-------------+   +-------------+
              |               |               |
              +---------------+---------------+
                              v
+---------------------------------------------------------------------+
|                   Phase 2: KB Search & Matching                     |
|  - Reference kb_index.json                                          |
|  - Primary KB prioritization                                        |
|  - Confidence score calculation                                     |
+---------------------------------------------------------------------+
                              |
              +---------------+---------------+
              v               v               v
    +-------------+   +-------------+   +-------------+
    | High Conf.  |   | Medium Conf.|   | Low Conf.   |
    |   (>=80%)   |   |  (50-79%)   |   |   (<50%)    |
    +-------------+   +-------------+   +-------------+
              |               |               |
              v               v               v
    +-------------+   +-------------+   +-------------+
    | Template 1  |   | Template 2  |   | Template 3  |
    | Solution    |   | Info Request|   | Escalation  |
    +-------------+   +-------------+   +-------------+
                              |
                              v
+---------------------------------------------------------------------+
|                     Phase 3: Response Draft Generation              |
|  - Template variable substitution                                   |
|  - KB steps integration                                             |
|  - Escalation determination                                         |
+---------------------------------------------------------------------+

Phase 1: Inquiry Analysis

1.1 Information Extraction

Extract from the ticket:

  • Customer name and contact info
  • Subject/title
  • Ticket number/ID
  • Full description
  • Previous correspondence (if any)

1.2 Auto-Detection Patterns

Configure patterns in references/kb_index.json:

PriorityPattern TypeExample RegexExample Match
1Error Code[Ee]rror\s*(\d{5})Error 30001
2Code Alternativecode\s*(\d{5})code 20002
3Device + Symptom(device_name)\s*(not working|offline)printer not working
4Hardware Symptom(not printing|won't boot|blue screen)not printing

1.3 Problem Category Identification

Map detected patterns to KB categories:

CategoryDetection CriteriaTypical KB
Error CodesNumeric error patternsError reference guides
Hardware IssuesPhysical device symptomsTroubleshooting guides
Software IssuesApplication errorsSoftware guides
Account/AccessLogin, permissionsAccount management
ConnectivityNetwork, offlineNetwork troubleshooting

Phase 2: KB Search and Confidence Scoring

2.1 Confidence Score Calculation

Match TypeScoreExample
Error code exact match+50Error 30001 in KB
Device/product name match+20"printer" mentioned
Keyword match+15"not printing"
Symptom similarity+10Similar symptom pattern
Category match+5Hardware category

2.2 Confidence Thresholds

LevelScoreAction
High>= 80Provide solution directly
Medium50-79Request additional information
Low< 50Escalate for investigation

2.3 Multiple KB Match Priority

When multiple KB articles match:

  1. Use primary_kb field if specified
  2. Prefer detailed_kb: true articles
  3. Prefer specific articles over general references

Phase 3: Response Draft Generation

3.1 Template Variables

VariableDescriptionSource
{{customer_name}}Customer nameTicket contact
{{ticket_subject}}Subject lineTicket subject
{{ticket_number}}Ticket IDTicket system
{{error_code}}Error codePattern detection
{{device_type}}Device/product typePattern detection
{{kb_article}}KB article nameKB match
{{kb_steps}}Troubleshooting stepsKB content
{{resolution_time}}Estimated timeKB metadata
{{escalation_reason}}Escalation reasonAnalysis result

Response Templates

Template 1: Clear Solution Available (High Confidence)

English:

Dear {{customer_name}},

Thank you for contacting our support team regarding {{ticket_subject}}.

Based on the information provided, this appears to be a {{device_type}} {{error_code}} issue. Here are the steps to resolve this:

**Troubleshooting Steps:**
{{kb_steps}}

**Estimated Resolution Time:** {{resolution_time}}

If the issue persists after completing these steps, please let us know and we will escalate to our technical team.

Best regards,
Support Team

Japanese:

{{customer_name}} 様

サポートへお問い合わせいただきありがとうございます。
「{{ticket_subject}}」についてご連絡いただきました。

ご報告いただいた内容から、{{device_type}}の{{error_code}}エラーと思われます。
以下の手順で解決できる可能性がございます:

**トラブルシューティング手順:**
{{kb_steps}}

**想定解決時間:** {{resolution_time}}

上記の手順でも解決しない場合は、技術チームへエスカレーションいたします。

サポートチーム

Template 2: Additional Information Needed (Medium Confidence)

English:

Dear {{customer_name}},

Thank you for contacting our support team regarding {{ticket_subject}}.

To better assist you with this issue, could you please provide the following information:

1. What is the exact error message or code displayed?
2. Which device/product is affected?
3. When did this issue first occur?
4. Have you tried any troubleshooting steps?
5. Has anything changed recently (updates, configuration)?

Once we receive this information, we will provide specific troubleshooting steps.

Best regards,
Support Team

Japanese:

{{customer_name}} 様

サポートへお問い合わせいただきありがとうございます。
「{{ticket_subject}}」についてご連絡いただきました。

より適切なサポートをご提供するため、以下の情報をお知らせください:

1. 表示されているエラーメッセージやコードは何ですか?
2. どの製品/デバイスで発生していますか?
3. この問題はいつ頃から発生していますか?
4. トラブルシューティングは試されましたか?
5. 最近何か変更はありましたか(更新、設定変更など)?

情報をいただき次第、具体的な対応方法をご案内いたします。

サポートチーム

Template 3: Escalation Required (Low Confidence)

English:

Dear {{customer_name}},

Thank you for contacting our support team regarding {{ticket_subject}}.

We have reviewed your request and determined that this issue requires detailed investigation by our technical team.

**Escalation Reason:** {{escalation_reason}}

We will:
1. Analyze the system logs and configuration
2. Coordinate with relevant teams if needed
3. Provide you with an update within [SLA timeframe]

In the meantime, please note any additional symptoms that may appear.

We appreciate your patience.

Best regards,
Support Team

Japanese:

{{customer_name}} 様

サポートへお問い合わせいただきありがとうございます。
「{{ticket_subject}}」についてご連絡いただきました。

ご報告いただいた内容を確認した結果、技術チームによる詳細調査が必要と判断いたしました。

**エスカレーション理由:** {{escalation_reason}}

以下の対応を進めます:
1. システムログと設定の分析
2. 必要に応じて関連チームとの連携
3. [SLA時間]以内に進捗をご報告

その間、追加の症状があればお知らせください。

ご理解のほどよろしくお願いいたします。

サポートチーム

Escalation Criteria

Automatic Escalation Conditions

ConditionThresholdAction
Repeat same error3+ times in 7 daysRequest pattern analysis
Unknown error codeNot in KB indexRecommend new KB creation
Multi-system failure2+ systemsImmediate escalation
Hardware replacementKB determinationVendor coordination

Priority Escalation Factors

Configure based on your business:

  • VIP customers/locations
  • High-traffic time periods
  • Weekend/holiday handling
  • SLA requirements

Escalation Handoff Format

When escalating to next-level support:

## Escalation Request

**Ticket Number:** #{{ticket_number}}
**Customer:** {{customer_name}}
**Issue Summary:** {{ticket_subject}}
**Confidence Score:** {{confidence_score}}

### Detected Information
- Error Code: {{error_code}}
- Device: {{device_type}}
- Symptoms: {{detected_keywords}}

### First Response Actions
- Referenced KB: {{kb_article}}
- Match Reason: {{match_reason}}
- Why Not Resolved: {{non_match_reason}}

### Escalation Reason
{{escalation_reason}}

### Recommended Investigation
1. {{investigation_item_1}}
2. {{investigation_item_2}}
3. {{investigation_item_3}}

### Related Information
- Similar past tickets: {{related_tickets}}
- Vendor contact needed: {{vendor_contact_needed}}

KB Index Configuration

Create a kb_index.json file with this structure:

{
  "version": "1.0",
  "kb_base_path": "./kb_articles",

  "auto_detection_patterns": {
    "error_code": {
      "pattern": "[Ee]rror\\s*(\\d+)",
      "priority": 1
    }
  },

  "confidence_thresholds": {
    "high": {"min_score": 80},
    "medium": {"min_score": 50, "max_score": 79},
    "low": {"min_score": 0, "max_score": 49}
  },

  "error_codes": {
    "10001": {
      "kb": "error_reference.md",
      "section": "Network Errors",
      "description": "Network connection error",
      "confidence": "high",
      "resolution_time": "5-15min"
    }
  },

  "keywords": {
    "printer": {
      "kb": ["printer_troubleshooting.md"],
      "aliases": ["printing", "print"],
      "confidence": "high",
      "category": "hardware"
    }
  },

  "categories": {
    "hardware": {
      "kb": ["hardware_guide.md"],
      "typical_resolution_time": "15-30min"
    }
  }
}

See references/kb_schema.json for the complete schema.


Usage Instructions

Basic Usage

  1. With ticket ID:

    /helpdesk-responder Ticket #1234 response draft
    
  2. With pasted content:

    /helpdesk-responder
    Please respond to:
    [paste email content]
    
  3. With specific error:

    /helpdesk-responder Error code 30001 troubleshooting
    
  4. In Japanese:

    /helpdesk-responder チケット#1234 日本語で回答
    

Output Format

The skill outputs:

  1. Problem Analysis

    • Detected error codes
    • Identified devices/products
    • Symptom keywords
    • Confidence score
  2. Referenced KB

    • Matched KB articles
    • Relevant sections
    • Solution applicability
  3. Response Draft

    • Template applied
    • Copy-paste ready
    • Language option
  4. Additional Actions

    • Follow-up questions (if info needed)
    • Escalation info (if required)
    • Vendor contacts (if applicable)

Customization Guide

Adapting to Your Organization

  1. Create KB Articles

    • Write troubleshooting guides in Markdown
    • Use consistent structure
  2. Configure kb_index.json

    • Map error codes to KB articles
    • Define keywords and aliases
    • Set confidence thresholds
  3. Customize Templates

    • Update company name
    • Adjust tone and style
    • Add signature blocks
  4. Set Escalation Rules

    • Define VIP criteria
    • Configure SLA timeframes
    • List vendor contacts

Version History

VersionDateChanges
1.02025-12-28Initial generic version

レビュー

まだレビューはありません。使ってみた感想をお寄せください。

同じリポジトリのスキル

概要と使いどころ

action-status-updater

無料日本語概要

Track and update action item status from natural language updates like 'Seanのメールには返信しておいた' or 'Lu対応予定'. Integrates with daily-comms-ops workflow. Use when managing action items across communication channels.

takusaotome/claude-skills-library92026年10月5日 更新

ai-adoption-consultant

無料日本語概要

AI活用提案の専門コンサルタントスキル。LLM・AIエージェントの活用事例知識を基に、業界別、部門別、シーン別に最適なAI導入提案を行います。システム化、業務改善、効率化を得意とし、具体的な活用事例、期待効果、導入ステップを含む実践的な提案を提供します。Use when proposing AI/LLM adoption strategies for specific industries, departments, or business scenarios.

takusaotome/claude-skills-library92026年10月5日 更新

Generate AI-powered BPO service proposals for Japanese companies in the US market. Use when creating outsourcing proposals with AI service menus, ROI estimation, implementation roadmaps, and bilingual (JA/EN) proposal documents for 在米日系企業向けAI実装.

日本語の概要は準備中です。原文の説明を表示しています。

takusaotome/claude-skills-library92026年10月5日 更新

ai-text-humanizer

無料日本語概要

AI(LLM)が生成した日本語テキストの「AI臭」を検出・診断し、人間らしい文章にリライトするスキル。 Use when: 「AIっぽい文章を直して」「人間らしくリライトして」「AI臭を消して」「この文章をもっと自然にして」 「テキストのAI感を減らして」「機械っぽさを取りたい」 "make this sound more human", "remove AI tone", "humanize this text", "detect if this is AI-generated". 6つのAI特有パターン(視覚的マーカー、単調なリズム、マニュアル的構成、非コミット姿勢、抽象語の濫用、 定型メタファー)を正規表現ベースで検出し、0-100のAI臭スコアを算出。3つの人間化技法 (バランスを崩す・客観を崩す・論理を崩す)でリライトを実行する。 Note: 検出スクリプトは日本語テキスト専用。英語テキストの場合はClaude自身がreferences/を参照して分析・リライトする。

takusaotome/claude-skills-library92026年10月5日 更新

Generate audit-ready internal control design documents from As-Is business process inventories. Produces control IDs, assertion mappings, procedures, SoD analysis, KPIs, materiality thresholds, and implementation roadmaps. Use when building internal controls for new audit engagements, SOX/J-SOX compliance, or process improvement initiatives.

日本語の概要は準備中です。原文の説明を表示しています。

takusaotome/claude-skills-library92026年10月5日 更新

Review audit-related documents (control design documents, bottleneck analyses, requirements definitions, etc.) for quality, scoring them 0-100 with a severity-rated findings list. Use when reviewing audit documents, checking control design quality, or verifying cross-document consistency. Supports documents governed by US GAAP, IFRS, or J-GAAP.

日本語の概要は準備中です。原文の説明を表示しています。

takusaotome/claude-skills-library92026年10月5日 更新

takusaotome のスキルをすべて見る

このスキルの問題を報告する