reply-classifier

Classify Gmail replies into categories and return structured JSON actions.

Updated Apr 19, 2026

Search Tags:#automation #workflow #ai #gmail #classification #triage #outreach

Ready to run

npx skills add https://github.com/marmikcfc/pepper-skills --skill reply-classifier-marmikcfc

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: reply-classifier
Source: https://github.com/marmikcfc/pepper-skills/tree/main/skills/reply-classifier
Command: npx skills add https://github.com/marmikcfc/pepper-skills --skill reply-classifier-marmikcfc

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Read recent Gmail replies from outreach sequences, classify each one using an LLM, and surface the right next action per category. Prevents interested leads from going cold, keeps do-not-contact lists accurate, and automates the triage work that usually falls through the cracks.

Core Features & Use Cases

Read unread Gmail replies from outreach sequences, classify each reply into one of the categories: interested, ooo, unsubscribe, objection, referral, other.
Surface the suggested next action per category and support downstream automation (e.g., draft a meeting invite, requeue, or update do-not-contact lists).
Provide a structured summary suitable for dashboards and pipelines.

Quick Start

Connect your Gmail account and run the classify workflow to triage unread replies.

Frequently Asked Questions about reply-classifier

High-intent search queries and answers about installing and using this skill.

FAQPage Schema

How do I automate Gmail reply triage for outreach workflows?▼

To automate Gmail reply triage, this Skill reads unread replies from outreach sequences and uses an LLM-based classifier to categorize them into interested, ooo, unsubscribe, objection, referral, or other. It returns structured JSON with category, confidence, reason, and suggested_action for each reply.

What is the best way to classify unsubscribe and out-of-office email replies?▼

The best way to classify unsubscribe and out-of-office email replies is using an LLM-based classifier that automatically reads unread Gmail responses and assigns categories. It surfaces a suggested next action per category, such as updating do-not-contact lists for unsubscribe replies or requeuing for ooo.

How does AI email classification route interested leads in outreach sequences?▼

AI email classification routes interested leads by using an LLM to evaluate unread Gmail replies and return a structured JSON output containing the category, confidence score, reason, and suggested_action. This prevents interested leads from going cold by supporting downstream automation like drafting a meeting invite.

Can I use this Gmail reply classifier for daily inbox triage without manual setup?▼

You can use this Gmail reply classifier for daily inbox triage by connecting your Gmail account and running the classify workflow. It requires no manual rule configuration for daily triage because the LLM-based classifier automatically evaluates replies and surfaces the right next action per category.

What structured data format does the Gmail reply classifier output for dashboards?▼

The Gmail reply classifier outputs a structured JSON format containing category, confidence, reason, and suggested_action for each reply. This structured summary is suitable for integration into dashboards and downstream automation pipelines to maintain do-not-contact lists and route interested leads.