MCP HubMCP Hub
返回技能列表

customer-brief

christopheryeo
更新于 3 days ago
27 次查看
1
在 GitHub 上查看
general

关于

This skill generates an executive-ready summary for a specific customer, combining internal communications, upcoming meetings, and external market intelligence. It is designed for pre-call preparation, outputting key bullets, recent touchpoints, and actionable insights. Key features include delivering structured outputs with sentiment scores, prep notes, and an audit trail of sources via the `list-files` skill.

快速安装

Claude Code

推荐
插件命令推荐
/plugin add https://github.com/christopheryeo/claude-skills
Git 克隆备选方式
git clone https://github.com/christopheryeo/claude-skills.git ~/.claude/skills/customer-brief

在 Claude Code 中复制并粘贴此命令以安装该技能

技能文档

Customer Brief

Deliver an audit-friendly briefing that blends current relationship signals, open work, and market context so account teams are ready for any touchpoint.

When to Use

  • Pre-call preparation for account directors, CSMs, or executive sponsors.
  • Requests such as "Run a customer brief for [Account]" or "What should I know before meeting [Account]?".

Success Criteria

  • 3–5 executive bullets with sentiment and next actions.
  • ≤7 recent communications with channel, parties, action, and confidence score.
  • ≤7 upcoming meetings normalized to the account timezone with prep notes.
  • Deliverables rendered via the list-files skill output block, highlighting overdue items.
  • ≥3 external insights (or explicit "No new items in last 14 days") with URLs and an audit trail of sources and keywords.

Guardrails

  • Do: Separate internal vs external signals, annotate owners/dates, log connectors and keywords.
  • Avoid: CRM data, gated sources, or summarizing private chats without user-provided exports.
  • Default research window: last 14 days (call out exceptions). Social scanning must stay public.
  • Keep every section tight—use bullets, sentence fragments ≤30 words, and remove repeated phrasing to minimize token usage.

Fast Prep Checklist

  1. Confirm account name/ID and primary timezone.
  2. Verify access to tagged email folders, relevant calendars, shared drives, and trackers.
  3. Align search plan (keywords, filters, priority sources) and sentiment rubric; start the audit log.

Workflow

  1. Internal Communications – Pull last 14 days of key emails/messages. Capture date (account TZ), channel, participants, ≤40 word summary, sentiment + confidence, and required action. State fallback if empty.
  2. Upcoming Touchpoints – Query calendar for next meetings. Normalize times, list organizer, attendees, prep tasks, attachments, and flag critical sessions.
  3. Deliverables & Files – Invoke list-files with the customer scope (folder/search query, limit ≤7, short summaries). Embed the returned # 📁 DRIVE FILE LISTING section as-is and note overdue statuses in follow-up bullets.
  4. Tasks/Trackers – Import open items with owner, due date, status, blockers. If unavailable, log the gap and request updates.
  5. External Intel – Run targeted web/news searches per plan. Capture up to five items with source, publish date, headline, and 1–2 sentence impact. Include search keywords, filters, and timestamp in audit log.
  6. Risks & Opportunities – Distill recurring themes into Risks, Opportunities, Watchlist with owners and next steps. Highlight sentiment swings or repeated overdue work.
  7. Executive Summary & Audit Log – Draft 3–5 bullets referencing supporting sections, maintain confident/empathetic tone, and append a concise audit log (sources, keywords, timestamps, manual overrides).

Output Skeleton

---
# Customer Brief: {{Account Name}}
*Generated: {{Local Timestamp}} | Primary Timezone: {{TZ}}*

## Executive Summary
- {{Insight + sentiment + action}}
- {{Insight}}
- {{Insight}}

## Recent Communications (Last 14 Days)
| Date ({{TZ}}) | Channel | Parties | Summary | Sentiment | Action |
| --- | --- | --- | --- | --- | --- |
| {{Sample row}} |

_Fallback_: "No new communications in last 14 days. Review archive label: {{Label}}."

## Upcoming Touchpoints
| Date | Time ({{TZ}}) | Meeting | Owner | Prep Notes |
| --- | --- | --- | --- | --- |
| {{Sample row}} |

{{Embed the `list-files` output here}}

## Risks & Opportunities
- **Risk:** {{Summary}} (Owner: {{Name}})
- **Opportunity:** {{Summary}} (Owner: {{Name}})
- **Watchlist:** {{Summary}} (Owner: {{Name}})

## External Highlights
1. _{{Source}}_ ({{Date}}) — {{Headline}} — {{Impact + action}} — [URL]
2. {{Additional items or "No new items in last 14 days."}}

## Search Log & Audit Notes
- Keywords: {{List}}
- Filters: {{Date range, types}}
- Sources Queried: {{List}}
- Exceptions: {{Access issues or manual inputs}}

Quality Gate

  • ≤7 rows per table unless user approves more.
  • Every external item cites a working URL and date.
  • Overdue/negative items flagged with owner follow-up.
  • Audit log includes connectors, keywords, timestamps, and manual overrides.

Validation Tips

  • Smoke-test with active vs dormant accounts.
  • Double-check timezone conversions on calendar entries.
  • Ensure communications summaries stay ≤40 words and sentiment labels include confidence.

GitHub 仓库

christopheryeo/claude-skills
路径: customer-brief

相关推荐技能

algorithmic-art

该Skill使用p5.js创建包含种子随机性和交互参数探索的算法艺术,适用于生成艺术、流场或粒子系统等需求。它能自动生成算法哲学文档(.md)和对应的交互式艺术代码(.html/.js),确保作品原创性避免侵权。开发者可通过定义计算美学理念快速获得可交互的艺术实现方案。

查看技能

subagent-driven-development

开发

该Skill用于在当前会话中执行包含独立任务的实施计划,它会为每个任务分派一个全新的子代理并在任务间进行代码审查。这种"全新子代理+任务间审查"的模式既能保障代码质量,又能实现快速迭代。适合需要在当前会话中连续执行独立任务,并希望在每个任务后都有质量把关的开发场景。

查看技能

executing-plans

设计

该Skill用于当开发者提供完整实施计划时,以受控批次方式执行代码实现。它会先审阅计划并提出疑问,然后分批次执行任务(默认每批3个任务),并在批次间暂停等待审查。关键特性包括分批次执行、内置检查点和架构师审查机制,确保复杂系统实现的可控性。

查看技能

cost-optimization

其他

这个Claude Skill帮助开发者优化云成本,通过资源调整、标记策略和预留实例来降低AWS、Azure和GCP的开支。它适用于减少云支出、分析基础设施成本或实施成本治理策略的场景。关键功能包括提供成本可视化、资源规模调整指导和定价模型优化建议。

查看技能