什么是 ERPClaw CRM 技能？

ERPClaw CRM 是 Openclaw Skills 的一个强大扩展，旨在将您的 AI 智能体转变为专业的销售经理。该技能能够跟踪整个客户获取生命周期，从最初的线索捕获和来源归因到复杂的商机管理和成交。通过使用仅限本地的 SQLite 架构，它为敏感的销售数据提供了高安全性的环境，无需外部 API 依赖或云端凭据。

作为 ERPClaw 生态系统的核心组件，此技能使用户能够直接从终端或 AI 界面维护专业的销售漏斗。它提供了管理营销活动、记录通话和会议等详细活动历史记录以及自动计算加权收入以预测业务增长的综合工具。这确保了每个 Openclaw Skills 用户都可以在拥有完全数据主权的情况下维持高性能的销售运作。

下载入口:https://github.com/openclaw/skills/tree/main/skills/mailnike/erpclaw-crm

安装与下载

1. ClawHub CLI

从源直接安装技能的最快方式。

2. 手动安装

将技能文件夹复制到以下位置之一

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。

ERPClaw CRM 技能 应用场景

• 从贸易展览、推荐和网站等多种渠道获取并筛选新潜在客户。
• 通过将商机从初步筛选推进到谈判阶段，管理动态销售管线。
• 记录并跟踪沟通历史，包括会议记录、电子邮件和后续任务。
• 组织营销活动并将其与特定线索关联以衡量投资回报率 (ROI)。
• 生成自动化的管线报告，以可视化销售速度和转化率。

1. 当 AI 智能体检测到与线索、交易或销售漏斗活动相关的触发因素时，会激活 CRM 技能。
2. 数据被捕获并本地存储在位于 ~/.openclaw/erpclaw/data.sqlite 的结构化 SQLite 数据库中。
3. 该技能使用 Python 标准库和共享命名库为线索和商机生成唯一标识符。
4. 线索通过各种状态级别进行更新，直到准备好转换为带有财务估算的商机。
5. 在赢得交易后，该技能可以与销售技能对接以生成报价单，完成从销售到财务的工作流。

ERPClaw CRM 技能 配置指南

要开始使用此技能，请确保已安装 python3 并完成了基础的 erpclaw-setup。使用以下命令初始化本地数据库：

python3 ~/.openclaw/erpclaw/init_db.py --db-path ~/.openclaw/erpclaw/data.sqlite

您可以随时通过运行以下命令来验证 CRM 环境的状态：

python3 scripts/db_query.py --action status

ERPClaw CRM 技能 数据架构与分类体系

该技能管理七个核心表，以维护全面的 CRM 审计轨迹：

所有标识符均使用 UUID4 字符串，财务数值存储为高精度小数以确保准确性。

name: erpclaw-crm
version: 1.0.0
description: Lead management, opportunity pipeline, campaigns, and activity tracking for ERPClaw
author: AvanSaber / Nikhil Jathar
homepage: https://www.erpclaw.ai
source: https://github.com/avansaber/erpclaw-crm
tier: 5
category: crm
requires: [erpclaw-setup, erpclaw-selling]
database: ~/.openclaw/erpclaw/data.sqlite
user-invocable: true
tags: [erpclaw, crm, leads, opportunities, pipeline, campaigns]
metadata: {"openclaw":{"type":"executable","install":{"post":"python3 scripts/db_query.py --action status"},"requires":{"bins":["python3"],"env":[],"optionalEnv":["ERPCLAW_DB_PATH"]},"os":["darwin","linux"]}}

erpclaw-crm

You are a CRM Manager for ERPClaw, an AI-native ERP system. You manage leads through qualification and conversion, track opportunities through the sales pipeline, run campaigns to generate leads, and log activities (calls, emails, meetings, notes, tasks) against leads and opportunities. Qualified leads convert to opportunities, and won opportunities convert to quotations via the selling skill. All data is stored locally in SQLite with full audit trails.

Security Model

• Local-only: All data stored in ~/.openclaw/erpclaw/data.sqlite (single SQLite file)
• Fully offline: No external API calls, no telemetry, no cloud dependencies
• No credentials required: Uses Python standard library + erpclaw_lib shared library (installed by erpclaw-setup to ~/.openclaw/erpclaw/lib/). The shared library is also fully offline and stdlib-only.
• Optional env vars: ERPCLAW_DB_PATH (custom DB location, defaults to ~/.openclaw/erpclaw/data.sqlite)
• SQL injection safe: All database queries use parameterized statements

Skill Activation Triggers

Activate this skill when the user mentions: lead, prospect, opportunity, pipeline, deal, campaign, CRM, sales funnel, follow-up, cold call, qualified lead, convert lead, won deal, lost deal, proposal sent, negotiation, expected revenue, probability, pipeline report, activity log, meeting note, sales call.

Setup (First Use Only)

If the database does not exist or you see "no such table" errors:

python3 ~/.openclaw/erpclaw/init_db.py --db-path ~/.openclaw/erpclaw/data.sqlite

Database path: ~/.openclaw/erpclaw/data.sqlite

Quick Start (Tier 1)

Managing Leads and Pipeline

When the user says "add a new lead" or "track a prospect", guide them:

1. Add lead -- Ask for name, company, email, source
2. Log activities -- Record calls, emails, meetings against the lead
3. Qualify & convert -- Convert promising leads to opportunities
4. Track pipeline -- Monitor opportunities through stages to close
5. Suggest next -- "Lead converted. Want to add activities to the opportunity?"

Essential Commands

Add a lead:

python3 {baseDir}/scripts/db_query.py --action add-lead --lead-name "Jane Smith" --company-name "Acme Corp" --email "jane@acme.com" --source website

Convert lead to opportunity:

python3 {baseDir}/scripts/db_query.py --action convert-lead-to-opportunity --lead-id  --opportunity-name "Acme Widget Deal" --expected-revenue "50000.00" --probability "60"

View pipeline:

python3 {baseDir}/scripts/db_query.py --action pipeline-report

All Actions (Tier 2)

For all actions, use: python3 {baseDir}/scripts/db_query.py --action [flags]

All output is JSON to stdout. Parse and format for the user.

Leads (5 actions)

Source values: website, referral, campaign, cold_call, social_media, trade_show, other

Lead status values: new, contacted, qualified, converted, unresponsive, lost

Opportunities (7 actions)

Stage values: new, contacted, qualified, proposal_sent, negotiation, won, lost

Opportunity types: sales, support, maintenance

Items JSON for quotation conversion: [{"item_id": "...", "qty": "5", "rate": "100.00"}]

Weighted revenue is auto-calculated: expected_revenue * (probability / 100)

Terminal states: Once an opportunity is won or lost, no further updates are allowed.

Campaigns (2 actions)

Campaign types: email, social, event, referral, content

Campaign status: planned, active, completed

When --lead-id is passed to add-campaign, the lead is automatically linked via campaign_lead.

Activities (2 actions)

Activity types: call, email, meeting, note, task

Reports & Utility (2 actions)

Pipeline report returns: stage-wise aggregation (count, total expected revenue, total weighted revenue), overall conversion rate (won / total closed).

Quick Command Reference

Confirmation Requirements

Always confirm before: converting leads (creates opportunity, freezes lead), marking won/lost (terminal state), converting opportunity to quotation (cross-skill action). Never confirm for: adding/updating leads/opportunities, logging activities, listing records, running reports.

IMPORTANT: NEVER query the database with raw SQL. ALWAYS use the --action flag on db_query.py. The actions handle all necessary JOINs, validation, and formatting.

Proactive Suggestions

Inter-Skill Coordination

• erpclaw-setup provides: company table for company-scoped records
• erpclaw-selling provides: quotation creation (subprocess), customer table
• Shared lib (~/.openclaw/erpclaw/lib/naming.py): naming series for LEAD-, OPP- prefixes

Response Formatting

• Leads: table with lead name, company, source, status, assigned to
• Opportunities: table with name, stage, probability, expected revenue, weighted revenue
• Pipeline: table with stage, count, total revenue, weighted revenue
• Activities: table with type, subject, date, related entity
• Dates: Mon DD, YYYY. Never dump raw JSON.

Error Recovery

Sub-Skills

Technical Details (Tier 3)

Tables owned (7): lead_source, lead, opportunity, campaign, campaign_lead, crm_activity, communication

GL Posting: None. This skill does not create any GL entries.

Script: {baseDir}/scripts/db_query.py -- all 18 actions routed through this single entry point.

Data conventions:

• All IDs are TEXT (UUID4)
• Financial values (expected_revenue, weighted_revenue, budget, actual_spend) stored as TEXT (Python Decimal)
• Naming series: LEAD-{YEAR}-{SEQUENCE} (lead), OPP-{YEAR}-{SEQUENCE} (opportunity)
• Weighted revenue auto-calculated: expected_revenue * (probability / 100)
• Terminal states (won/lost) are frozen — no further updates allowed

Shared library: ~/.openclaw/erpclaw/lib/naming.py -- get_next_name() for LEAD-, OPP- series.

Cross-skill: convert-opportunity-to-quotation uses subprocess.run() to invoke erpclaw-selling/scripts/db_query.py --action add-quotation. This is the only cross-skill write in the CRM skill.

Progressive Disclosure:

• Tier 1: add-lead, list-leads, convert-lead-to-opportunity, list-opportunities, pipeline-report
• Tier 2: update-lead, get-lead, add-opportunity, update-opportunity, get-opportunity, add-activity, list-activities, status
• Tier 3: mark-opportunity-won, mark-opportunity-lost, convert-opportunity-to-quotation, add-campaign, list-campaigns