← 返回 Skills 市场
ceobotson-bot

DoctorClaw Client Follow-Up

作者 ceobotson-bot · GitHub ↗ · v1.0.0 · MIT-0
cross-platform ⚠ suspicious
311
总下载
0
收藏
0
当前安装
1
版本数
在 OpenClaw 中安装
/install doctorclaw-client-follow-up
功能描述
Automated client follow-up — scan your client list, find stale leads, draft personalized outreach. Weekly cron or on-demand.
使用说明 (SKILL.md)

Automated Client Follow-Up

Stop losing deals to silence. This skill reads your client or lead list, identifies contacts you haven't reached out to recently, and drafts personalized follow-up messages — so no opportunity slips through the cracks.

Run it weekly on a cron, or trigger it whenever you want to check who needs attention.

What You Get

  • Stale lead detection — contacts you haven't messaged in X days
  • Personalized follow-up drafts for each stale contact
  • Priority ranking based on deal value, last interaction, and lead stage
  • Follow-up stats (total contacts, stale count, overdue count)
  • Outreach log tracking — what you sent, when, and to whom

Setup

Required

  • Client list — A CSV file, Google Sheet, or any structured list your agent can read. Minimum columns: name, email (or phone), last-contacted date. More columns = better personalization.

Optional (but recommended)

  • Send access — If you want the agent to send follow-ups after your approval (Gmail, Outlook, or SMS via Twilio)
  • CRM integration — If you use a CRM (HubSpot, Notion database, Airtable), point the agent at it instead of a CSV
  • Calendar access — So the agent can suggest meeting times in follow-ups
  • Delivery channel — Telegram/Discord for follow-up digest notifications

Configuration

Tell your agent:

  1. Client list location — file path, Google Sheet URL, or CRM connection
  2. Stale threshold — how many days without contact before a lead is "stale" (default: 14 days)
  3. Critical threshold — how many days before a lead is "at risk" (default: 30 days)
  4. Follow-up style — your tone (professional, casual, friendly, direct) so drafts match your voice
  5. Follow-up schedule — when to run (default: every Monday at 9:00 AM local)
  6. Max follow-ups per run — limit drafts per cycle (default: 10)
  7. Delivery — where to send the follow-up digest (Telegram, Discord, file)
  8. Industry/context — what your business does, so follow-ups are relevant (e.g., "web design agency", "real estate investor", "SaaS founder")

How It Works

Step 1: Load Client List

  • Read your client list from the configured source (CSV, Google Sheet, CRM, Notion)
  • For each contact, extract: name, email/phone, company (if available), last-contacted date, deal stage, deal value, notes
  • If last-contacted date is missing, flag the contact for manual review

Step 2: Identify Stale Contacts

Sort contacts into 3 categories based on days since last contact:

🔴 AT RISK — Over critical threshold (30+ days)

  • These leads are going cold. Follow-up is urgent.
  • Prioritize by deal value (highest value first)

🟡 STALE — Over stale threshold (14-30 days)

  • Due for a check-in. Not urgent, but shouldn't wait another week.
  • Prioritize by deal stage (closest to closing first)

🟢 ACTIVE — Within threshold (\x3C 14 days)

  • Recently contacted. No action needed this cycle.
  • Skip these — don't over-contact

Step 3: Rank & Select

  • Rank all stale + at-risk contacts by priority score:
    • Deal value weight: higher value = higher priority
    • Days overdue weight: more overdue = higher priority
    • Deal stage weight: closer to closing = higher priority
  • Select top N contacts (up to configured max per run)

Step 4: Draft Follow-Ups

For each selected contact, draft a personalized follow-up message:

  • Use context: reference their company, deal stage, last conversation topic (from notes)
  • Match tone: use the configured follow-up style
  • Keep it short: 3-5 sentences max
  • Include a clear CTA: ask a question, propose a meeting, share something useful
  • Vary the approach: don't send the same template to everyone
    • Re-engagement: "Haven't heard from you in a while — still interested in X?"
    • Value-add: "Saw this article about [their industry] and thought of you"
    • Check-in: "How's [project/initiative they mentioned] going?"
    • Nudge: "We had discussed [service/product] — any questions I can answer?"
    • Meeting request: "Would love to catch up — free for a quick call this week?"
  • Mark as DRAFT — never send without user approval

Step 5: Compile Follow-Up Digest

Format the digest:

🤝 Follow-Up Digest — [Date]

📊 PIPELINE STATUS
Total contacts: [X] | At risk: [X] | Stale: [X] | Active: [X]
Oldest untouched: [X days] — [Contact Name]

🔴 AT RISK ([X] contacts)
1. [Name] — [Company] | Last contact: [X days ago]
   Stage: [deal stage] | Value: [deal value]
   📝 Draft: "[First line of follow-up...]"

2. [Name] — [Company] | Last contact: [X days ago]
   Stage: [deal stage] | Value: [deal value]
   📝 Draft: "[First line of follow-up...]"

🟡 STALE ([X] contacts)
3. [Name] — [Company] | Last contact: [X days ago]
   Stage: [deal stage]
   📝 Draft: "[First line of follow-up...]"

4. [Name] — [Company] | Last contact: [X days ago]
   Stage: [deal stage]
   📝 Draft: "[First line of follow-up...]"

🟢 ACTIVE ([X] contacts) — no action needed

💡 INSIGHTS
• [X] contacts haven't been reached in 30+ days
• Top deal at risk: [Name] — $[value]
• Suggested: Block 30 min this week for follow-up calls

Step 6: Deliver & Track

  • Send digest via configured channel (Telegram, Discord, or save to file)
  • Show full draft messages below the digest for review
  • When user approves a follow-up ("send 1" or "approve all"), send the messages
  • After sending, update the client list with the new last-contacted date
  • Log the outreach in memory/follow-up-log/YYYY-MM-DD.md

Examples

User: "Check my follow-ups"

Agent runs the workflow and responds:

🤝 Follow-Up Digest — Monday, March 10

📊 PIPELINE STATUS Total contacts: 34 | At risk: 3 | Stale: 7 | Active: 24 Oldest untouched: 45 days — Lisa Park

🔴 AT RISK (3 contacts)

  1. Lisa Park — Greenfield Properties | Last contact: 45 days ago Stage: Proposal sent | Value: $4,800 📝 Draft: "Hi Lisa, I wanted to circle back on the proposal I sent over last month. Happy to walk through any questions or adjust the scope if needed. Would Thursday or Friday work for a quick call?"

  2. James Wu — TechStart Inc | Last contact: 38 days ago Stage: Discovery | Value: $2,400 📝 Draft: "Hey James, we had a great conversation about streamlining your onboarding process. I put together a few ideas since we last spoke — mind if I send them over?"

  3. Maria Santos — Bloom Studio | Last contact: 31 days ago Stage: Interested | Value: $1,200 📝 Draft: "Hi Maria, just checking in — are you still looking to revamp your client intake process? I'd love to help whenever the timing is right."

🟡 STALE (7 contacts) 4. Dave Chen — Redline Motors | 22 days ago 5. Priya Patel — Sapphire Consulting | 19 days ago 6. Tom Bradley — Bradley & Sons | 18 days ago 7. (+ 4 more)

🟢 ACTIVE (24 contacts) — no action needed

💡 INSIGHTS • 3 contacts at risk of going cold — $8,400 in pipeline value • Lisa Park is your highest-value stale lead • Suggested: Prioritize Lisa and James this week

User: "Send 1 and 2. Edit 3 to mention we have a spring discount."

Agent: Sends follow-ups 1 and 2 as drafted. Updates draft 3 with spring discount mention, shows updated version for approval.


User: "Set up weekly follow-ups every Monday"

Agent: Configures cron for Monday 9:00 AM, confirms:

"Client follow-up scheduled for every Monday at 9:00 AM. I'll scan your contact list, flag anyone going cold, and draft personalized messages for your review."


User: "Add a new contact: Sarah Kim, [email protected], met at conference, interested in the $495 setup"

Agent: Adds to client list with today as first-contact date, stage: "Interested", notes: "Met at conference, interested in Guided Setup ($495)." Confirms the addition.

Customization Ideas

  • Multi-channel follow-up — draft emails AND LinkedIn messages for the same contact
  • Auto-categorize new leads — when a new contact is added, auto-assign a deal stage and priority
  • Follow-up sequences — instead of one-off messages, create multi-touch sequences (Day 1: email, Day 3: LinkedIn, Day 7: call reminder)
  • Win/loss tracking — mark deals as won or lost, track conversion rate over time
  • Referral prompts — for active clients, suggest asking for referrals after successful projects
  • Meeting prep — before scheduled calls, pull up the contact's history and draft talking points

Want More?

This skill handles follow-up tracking and outreach drafting. But if you want:

  • Custom integrations — connect to your CRM, project management tool, invoicing system, or any API your business uses
  • Advanced automations — multi-step workflows tailored to your business (lead scoring, auto-replies, invoice follow-ups, proposal generation)
  • Full system setup — identity, memory, security, and 5 custom automations built specifically for your workflow

DoctorClaw sets up complete OpenClaw systems for businesses:

  • Guided Setup ($495) — 2-hour live walkthrough. Everything configured, integrated, and running by the end of the call.
  • Done-For-You ($1,995) — 7-day custom build. 5 automations, 3 integrations, full security, 30-day support. You do nothing except answer a short intake form.

doctorclaw.ceo

安全使用建议
This skill appears to do what it says, but be cautious when connecting real data sources and delivery channels: 1) Only point the agent at the exact CSV/Sheet/CRM you intend and avoid giving broad admin credentials—use read-only or narrowly scoped service accounts where possible. 2) If you enable sending (Gmail/Outlook/Twilio), validate that the agent will prompt for approval before sending (the skill says drafts are marked DRAFT and 'never send without user approval'). 3) Review and audit the follow-up-log files the agent writes, and confirm the agent's update behavior (it will update last-contacted dates). 4) If you want stronger guarantees, test with a small sample dataset and a non-production/test account before connecting full production contacts.
功能分析
Type: OpenClaw Skill Name: doctorclaw-client-follow-up Version: 1.0.0 The skill 'client-follow-up' (SKILL.md) requests broad, high-privilege access to sensitive data sources including CRMs (HubSpot, Airtable), local files (CSV), and communication channels (Email, SMS, Discord). While these capabilities are plausibly required for its stated purpose and the instructions include a safety check for user approval before sending messages, the skill lacks any input sanitization logic for processing untrusted data like 'client notes,' which creates a significant surface for indirect prompt injection. The inclusion of marketing links to an external service (doctorclaw.ceo) and the request for persistent execution via cron further elevate the risk profile.
能力评估
Purpose & Capability
The name and description (scan contacts, identify stale leads, draft follow-ups, optionally send them) align with the SKILL.md instructions. All actions described (reading a client list, ranking, drafting, delivering a digest, logging) are coherent for this purpose.
Instruction Scope
The SKILL.md focuses on reading a configured client list (CSV/Google Sheet/CRM/Notion) and drafting messages, which is on-purpose. It is a bit permissive/vague about how the agent should obtain access to remote CRMs or Google Sheets (e.g., 'point the agent at it'), so at runtime the agent may need credentials or access tokens you must explicitly provide. The skill instructs updating client lists and writing logs to memory/follow-up-log, which is within scope but means the agent will modify data sources you configure.
Install Mechanism
Instruction-only skill with no install spec and no code files. Nothing is downloaded or written by the bundle itself, minimizing install-time risk.
Credentials
The skill bundle declares no required environment variables or credentials, but the instructions reference optional integrations (Gmail, Outlook, Twilio, Google Sheets, HubSpot, Airtable, Notion, Telegram/Discord, Calendar) that will require credentials at runtime if you enable them. The lack of declared credentials in the bundle is not inherently dangerous, but you should only supply the minimal credentials needed and prefer scoped API keys/service accounts where possible.
Persistence & Privilege
always is false and the skill does not request permanent platform privileges. It does instruct writing follow-up logs and updating the configured client list (expected for this functionality). No indications it will modify other skills or global agent settings.
如何使用
  1. 确保已安装 OpenClaw(本地或 Docker 部署)
  2. 在对话框中输入安装命令:/install doctorclaw-client-follow-up
  3. 安装完成后,直接呼叫该 Skill 的名称或使用 /doctorclaw-client-follow-up 触发
  4. 根据 Skill 的参数说明提供必要输入,即可获得结构化输出
版本历史
v1.0.0
Initial release — automated client follow-up workflow to identify and engage stale leads. - Scans your client or lead list to detect contacts you haven’t reached out to recently. - Ranks and prioritizes follow-ups based on deal value, last interaction, and lead stage. - Drafts personalized follow-up messages (never sent without approval) and generates a follow-up digest. - Delivers summary digests via file or messaging channels; supports on-demand and weekly scheduled runs. - Logs outreach actions and updates last-contacted dates to keep your process organized. - Highly configurable: thresholds, follow-up style, delivery channel, list source (CSV, Google Sheets, CRM), and more.
元数据
Slug doctorclaw-client-follow-up
版本 1.0.0
许可证 MIT-0
累计安装 0
当前安装数 0
历史版本数 1
常见问题

DoctorClaw Client Follow-Up 是什么?

Automated client follow-up — scan your client list, find stale leads, draft personalized outreach. Weekly cron or on-demand. 它是一个面向 Claude Code / OpenClaw 的 AI Agent Skill 插件,目前累计下载 311 次。

如何安装 DoctorClaw Client Follow-Up?

在 OpenClaw 或 Claude Code 对话框中运行命令「/install doctorclaw-client-follow-up」即可一键安装,无需额外配置。

DoctorClaw Client Follow-Up 是免费的吗?

是的,DoctorClaw Client Follow-Up 完全免费,采用 MIT-0 许可证,可自由下载、安装和使用。

DoctorClaw Client Follow-Up 支持哪些平台?

DoctorClaw Client Follow-Up 跨平台运行,可在任意部署了 OpenClaw / Claude Code 的环境中使用(cross-platform)。

谁开发了 DoctorClaw Client Follow-Up?

由 ceobotson-bot(@ceobotson-bot)开发并维护,当前版本 v1.0.0。

💬 留言讨论