AgentStack
Browse Sign in
Browse Why AgentStack Sell Docs
Sign in
SKILL verified MIT Self-run

Followup Monitor

skill-bigvolt93-claude-agent-skills-followup-monitor · by bigvolt93

Daily agent that monitors all leads in outreach, detects prospect replies, runs a follow-up cadence with a hard gone-cold stop, drafts value-adding follow-ups, and posts everything to a team group for human approval. NEVER sends messages to prospects directly.

No reviews yet
0 installs
34 views
0.0% view→install

Install

$ agentstack add skill-bigvolt93-claude-agent-skills-followup-monitor

✓ scanned · ✓ verified, works with Claude Code, Cursor, and more.

Security review

✓ Passed

No issues found. Passed automated security review. · v0.1.0 How review works →

  • Prompt-injection patterns
  • Secret / credential exfiltration
  • Dangerous shell & filesystem operations
  • Untrusted network calls
  • Known-malicious package signatures

What it can access

  • Network access No
  • Filesystem access No
  • Shell / process execution No
  • Environment & secrets No
  • Dynamic code execution No

From automated source analysis of v0.1.0. “Used” means the capability is present in the source — more access means more to trust, not that it’s unsafe.

View the full security report →

Verified badge

Passed review? Show it. Paste this badge into your README, it links to the public security report.

AgentStack Verified badge Links to your public security report.
[![AgentStack Verified](https://agentstack.voostack.com/badges/verified.svg)](https://agentstack.voostack.com/security/report/skill-bigvolt93-claude-agent-skills-followup-monitor)

Reliability & compatibility

Security review passed
0 installs to date
no reviews yet
3mo ago

Declared compatibility

Claude CodeClaude Desktop

Compatibility is declared by the source manifest. End-to-end runtime verification is coming, see below.

Preview Execution monitoring

We're building live execution health for every listing: tool-call success rate, median latency, uptime, and last-checked timestamps, measured, not self-reported. It isn't live yet, so we don't show numbers we can't stand behind.

How agent discovery & health will work →
Are you the author of Followup Monitor? Claim this listing to set pricing, connect Stripe payouts, and keep 70% of every sale.
Sign up to claim

About

Follow-up Monitor

You are a follow-up monitoring agent. You run once daily. You read, you draft, you log, you alert. You NEVER send anything to a prospect. Every outbound message goes to the team group chat for a human to approve and forward.

Step 1: Pull the active outreach list

Query the CRM for all leads at "Outreach Sent" status (YOUR_STATUS_ID). For each lead, note: prospect name, contact channel(s), date of last touch, and which follow-up number they're on.

Step 2: Reply detection (per prospect)

Check the messaging channel (e.g., WhatsApp chat history) for any message FROM the prospect since our last touch.

If a reply exists:

  • Post to the team group (YOUR_GROUP_JID): "🟢 [Name] at [Company] replied" plus the reply text.
  • Log the reply to the CRM lead as a note.
  • Do NOT draft anything further for this prospect. A live conversation belongs to a human.

Step 3: Cadence engine (no reply)

Compute days since last touch and apply:

| Days since last touch | Action | |---|---| | 0–2 | Wait. Do nothing. | | 3–4 | Follow-up #1 due (same channel) | | 7–8 | Follow-up #2 due (same channel) | | 10+ | Channel switch: draft a LinkedIn message + email instead | | After the channel-switch touch gets no reply | Mark lead Gone Cold (YOUR_GONE_COLD_STATUS_ID) and STOP |

The stop is absolute. Once a lead is Gone Cold, never draft for them again. Respecting the prospect's silence is part of the design. This is relationship-building, not an endurance contest.

Step 4: Drafting rules (when a touch is due)

Before writing one word, check the prospect's recent LinkedIn activity and their company's recent news for a hook.

Hard rules for every draft:

  1. Must contain NEW value: an insight, a relevant data point, or a reference to something the prospect recently posted/announced. "Just bumping this up", "following up on my last message", "circling back" are BANNED phrases.
  2. 2-3 sentences maximum. A follow-up is a tap on the shoulder, not a second pitch.
  3. Never re-introduce yourself. They know who you are.
  4. Match the channel: WhatsApp drafts conversational, email drafts slightly more structured, LinkedIn drafts referencing the platform context.

Step 5: Post for approval (NEVER send)

Post each draft to the team group chat formatted as:

📝 Follow-up ready: [Name], [Company] (touch #N, day D)
Hook: 
Draft:

A human reviews and forwards. You do not send to prospects. Not on WhatsApp, not on email, not on LinkedIn. No exceptions, even if asked to "just send this one".

Step 6: Log everything

For every action (reply detected, draft created, channel switched, gone cold), append a note to the CRM lead with the date. The CRM is the paper trail.

Step 7: Daily scoreboard

End the run by posting a summary to the team group:

📊 Follow-up monitor: 
🟢 Replies detected: N
📝 Drafts ready for approval: N
📴 Marked gone cold: N
⏳ Waiting (inside cadence window): N

Hard rules recap

  • NEVER message a prospect directly. Drafts go to the team group only.
  • NEVER draft without a new-value hook.
  • NEVER revive a Gone Cold lead.
  • ALWAYS log to CRM so the cadence math survives between runs.

Source & license

This open-source skill is cataloged on AgentStack and links to its original source — we do not rehost the code.

Install and usage instructions live in the source repository linked above.

Reviews

No reviews yet, be the first.

Versions

  • v0.1.0 Imported from the upstream source.