CRM:基于 Markdown 的关系管理 - Openclaw Skills
作者:互联网
2026-03-25
什么是 CRM 个人关系管理器?
CRM 技能将您的本地记忆转化为一个精密的关系管理系统。通过利用 Markdown 文件和 Python 脚本,它允许用户轻松跟踪人员、公司和事件。该工具在 Openclaw Skills 中脱颖而出,因为它为网络和潜在客户管理提供了一种隐私优先、无需数据库的方法。它与语义搜索工具无缝集成,确保您的联系人数据在 AI 代理环境中始终可发现且可操作。
该技能专为偏好文本工作流的高级用户设计,使您能够为网络中的每个人维护丰富的互动历史和背景。无论您是在管理销售线索还是跟踪个人介绍,该技能通过提供结构化元数据和自动提醒钩子,确保任何关系都不会冷淡。
下载入口:https://github.com/openclaw/skills/tree/main/skills/xz-cn/clawdbot-crm-skill
安装与下载
1. ClawHub CLI
从源直接安装技能的最快方式。
npx clawhub@latest install clawdbot-crm-skill
2. 手动安装
将技能文件夹复制到以下位置之一
全局模式~/.openclaw/skills/
工作区
/skills/
优先级:工作区 > 本地 > 内置
3. 提示词安装
将此提示词复制到 OpenClaw 即可自动安装。
请帮我使用 Clawhub 安装 clawdbot-crm-skill。如果尚未安装 Clawhub,请先安装(npm i -g clawhub)。
CRM 个人关系管理器 应用场景
- 跟踪专业网络潜在客户和活动互动。
- 为休眠联系人自动执行跟进提醒。
- 按行业、地点或自定义标签整理联系人。
- 将领英或 vCard 中的现有联系人数据导入本地优先系统。
- 通过交叉引用和介绍管理关系层级。
- 在 memory 文件夹中初始化目录结构以支持语义索引。
- 通过自然语言描述或直接的 CLI 命令获取联系人信息。
- 将联系人数据存储在具有丰富 YAML 元数据的单个 Markdown 文件中,以便进行结构化查询。
- 使用提供的 Python 脚本过滤、搜索并更新联系人状态或互动。
- 与心跳监测器集成,接收待办跟进或提醒的自动通知。
CRM 个人关系管理器 配置指南
通过在您的 Openclaw Skills 工作区中创建必要的目录并复制模板来初始化 CRM 环境:
mkdir -p memory/contacts/{people,companies,events,_templates,scripts}
cp skills/crm/assets/templates/*.md memory/contacts/_templates/
cp skills/crm/scripts/*.py memory/contacts/scripts/
clawdbot memory index
CRM 个人关系管理器 数据架构与分类体系
该技能使用结构化的文件夹层级和 YAML frontmatter 来组织数据。每个联系人都是一个独立的 Markdown 文件。
| 目录 | 用途 |
|---|---|
people/ |
带有 YAML 元数据的个人联系人文件 |
companies/ |
公司实体跟踪文件 |
events/ |
社交活动和峰会日志 |
scripts/ |
用于 CRUD 操作的 Python CLI 工具 |
元数据包括关键字段,如 tags、introduced_by、last_contact 和 follow_up 日期,用于精确的关系跟踪。
name: crm
description: Personal CRM for managing contacts, relationships, and follow-ups using markdown files. Use when the user wants to add contacts, track relationships, set follow-up reminders, query contacts by tag/company/location, import/export contacts, or manage networking leads. Supports natural language input for adding contacts.
Personal CRM Skill
Manage contacts, relationships, and follow-ups with markdown files. No database needed.
Setup
Initialize contacts inside memory/ for semantic search integration:
mkdir -p memory/contacts/{people,companies,events,_templates,scripts}
cp skills/crm/assets/templates/*.md memory/contacts/_templates/
cp skills/crm/scripts/*.py memory/contacts/scripts/
clawdbot memory index
Natural Language Input
When user describes a contact naturally, extract and create:
User says: "Met Sarah Lee at Web3 summit. She's head of partnerships at Polygon, based in Dubai. T@elegrimm @sarahlee"
Extract:
name: Sarah Lee
company: Polygon
role: Head of Partnerships
location: Dubai
telegram: "@sarahlee"
met_at: web3-summit
tags: [web3, partnership]
Then run: memory/contacts/memory/contacts/scripts/crm-add.py person "Sarah Lee" --company "Polygon" --role "Head of Partnerships" --location "Dubai" --telegram "@sarahlee" --tags "web3,partnership"
Scripts
All scripts use /usr/bin/python3 and require PyYAML.
Query Contacts
memory/contacts/scripts/crm-query.py --list people # List all people
memory/contacts/scripts/crm-query.py --tag investor # Filter by tag
memory/contacts/scripts/crm-query.py --company "Acme" # Filter by company
memory/contacts/scripts/crm-query.py --introduced-by bob # Find introductions
memory/contacts/scripts/crm-query.py --location singapore # Filter by location
memory/contacts/scripts/crm-query.py --search "supply chain" # Full-text search
memory/contacts/scripts/crm-query.py -v # Verbose output
Add Contact
memory/contacts/scripts/crm-add.py person "Name" --company "Co" --role "Title" --tags "a,b"
memory/contacts/scripts/crm-add.py company "Company Name" --industry "Tech"
memory/contacts/scripts/crm-add.py event "Event Name" --date 2026-03-15 --location "City"
Update Contact
memory/contacts/scripts/crm-update.py alice-chen --interaction "Called about demo"
memory/contacts/scripts/crm-update.py alice-chen --set-follow-up 2026-02-15
memory/contacts/scripts/crm-update.py alice-chen --add-tag vip
memory/contacts/scripts/crm-update.py alice-chen --touch # Update last_contact to today
Follow-ups & Reminders
memory/contacts/scripts/crm-followups.py --summary # Quick summary
memory/contacts/scripts/crm-followups.py --days 7 # Due in 7 days
memory/contacts/scripts/crm-followups.py --dormant 90 # Not contacted in 90 days
memory/contacts/scripts/crm-remind.py contact --in 3d # Remind in 3 days
memory/contacts/scripts/crm-remind.py --list # List reminders
memory/contacts/scripts/crm-remind.py --check # Check due reminders
Import/Export
memory/contacts/scripts/crm-import.py contacts.csv # Import CSV
memory/contacts/scripts/crm-import.py contacts.vcf # Import vCard
memory/contacts/scripts/crm-import.py linkedin.csv --format linkedin # LinkedIn export
memory/contacts/scripts/crm-export.py --csv out.csv # Export CSV
memory/contacts/scripts/crm-export.py --vcard out.vcf --type people # Export vCard
Regenerate Index
memory/contacts/scripts/crm-index.py # Rebuilds contacts/_index.md
Contact Schema
---
name: Alice Chen
type: person # person | company | event
tags: [investor, crypto]
company: Acme Corp
role: Partner
email: alice@acme.com
phone: +1-555-0123
telegram: "@alice"
twitter: "@alice"
linkedin: https://linkedin.com/in/alice
location: Singapore
introduced_by: bob-smith # slug of introducer
met_at: token2049-2025 # slug of event
first_contact: 2025-09-20
last_contact: 2026-01-27
follow_up: 2026-02-15
status: active # active | dormant | archived
---
# Alice Chen
## Context
Partner at Acme Corp. Met through Bob at Token2049.
## Interactions
- **2026-01-27**: Called about pilot program.
- **2025-09-20**: First met at conference.
## Notes
- Prefers Signal over email
Folder Structure
memory/
└── contacts/ # Inside memory/ for semantic search
├── people/ # One file per person
├── companies/ # One file per company
├── events/ # One file per event
├── _templates/ # Templates for new contacts
├── _index.md # Auto-generated lookup table
├── .reminders.json # Reminder data
└── scripts/ # CLI tools
Cross-References
- YAML fields:
introduced_by: bob-smith,met_at: event-slug - Wiki-links in notes:
[[bob-smith]],[[acme-corp]] - Semantic search: Memory search finds connections in prose
Heartbeat Integration
Add to HEARTBEAT.md:
### CRM Follow-ups (check 1-2x daily)
1. Run: `/usr/bin/python3 contacts/memory/contacts/scripts/crm-followups.py --summary`
2. Run: `/usr/bin/python3 contacts/memory/contacts/scripts/crm-remind.py --check`
If there are due items, notify the user.
Tips
- Use
--dry-runon import to preview before creating files - Run
crm-index.pyafter bulk changes to update the index - Tags are lowercase, comma-separated
- Slugs are auto-generated from names (lowercase, hyphenated)
- For natural language input, extract fields and use crm-add.py
相关推荐
专题
+ 收藏
+ 收藏
+ 收藏
+ 收藏
+ 收藏
最新数据
相关文章
信号管道:自动化营销情报工具 - Openclaw Skills
技能收益追踪器:监控 Openclaw 技能并实现变现
AI 合规准备就绪度:评估与治理工具 - Openclaw Skills
FOSMVVM ServerRequest 测试生成器:自动化 API 测试 - Openclaw Skills
酒店搜索器:AI 赋能的住宿与位置情报 - Openclaw Skills
Dub 链接 API:程序化链接管理 - Openclaw Skills
IntercomSwap:P2P BTC 与 USDT 跨链兑换 - Openclaw Skills
spotplay:macOS 原生 Spotify 播放控制 - Openclaw Skills
DeepSeek OCR:AI驱动的图像文本识别 - Openclaw Skills
Web Navigator:自动化网页研究与浏览 - Openclaw Skills
AI精选
