cs-reply-drafter

Drafts customer service replies by checking stock, pricing, and credit status in AutoCount.

1|Updated Jun 21, 2026
One-click install
npx skills add https://github.com/tapway/shogun-os --skill cs-reply-drafter-tapway
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: cs-reply-drafter
Source: https://github.com/tapway/shogun-os/tree/main/skills/crm/cs-reply-drafter
Command: npx skills add https://github.com/tapway/shogun-os --skill cs-reply-drafter-tapway

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) components.

What problem does it solve? Customer service staff waste time manually checking stock levels, contract pricing, and credit status across systems before replying to enquiries, leading to slow responses and stale data. This Skill automates the data-gathering and drafting step so staff only review and approve. ## Core Features & Use Cases - Automated Data Lookups: Parses inbound enquiries, resolves customer identity, checks real-time stock via AutoCount, applies customer-specific pricing rules, and verifies credit status through accounting aging reports. - Template-Based Drafts: Selects from six reply templates (available, partial, unavailable, credit-hold, new-customer, unclear) and produces a structured JSON draft with line-item details and approval flags. - Batch Processing: Processes overnight enquiry queues in bulk and generates a batch summary with status breakdowns. - Use Case: A customer messages on WhatsApp asking for 10 units of Widget A. The Skill checks stock (7 available), pulls the contract price, confirms credit is clear, and produces a ready-to-review reply draft flagging the 3-unit shortfall with a backorder ETA. ## Quick Start Ask the agent to draft a reply for a customer enquiry about product availability and pricing using the cs-reply-drafter skill.

Frequently Asked Questions about cs-reply-drafter

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

FAQPage Schema
How do I draft customer service replies with stock and pricing data?

Run draft_reply.py with --enquiry-text or --enquiry-file to parse the enquiry, check AutoCount stock, apply pricing rules, and produce a JSON reply draft. Staff review the draft in the reply-drafts output directory before sending.

How do I process multiple customer enquiries in batch?

Use the --batch flag with --queue-dir pointing to a directory of enquiry JSON files. The script processes each enquiry, writes individual drafts, and generates a _batch_summary.json with counts by status.

Does this skill send replies to customers automatically?

No, it only prepares drafts for staff review. Sending is handled by CRM bridges like respondio-bridge or chatwoot-bridge, and drafts flagged as partial, unavailable, unclear, or credit-hold always require approval.

What happens when a customer enquiry cannot be parsed?

Low-confidence enquiries are flagged as unclear and routed to a human using the reply-unclear template. The skill never guesses a SKU, and items without identified SKUs are marked for manual lookup.

Can I test reply drafting without writing output files?

Yes, use the --dry-run flag to print drafts to stdout without writing to the reply-drafts directory. Missing or incomplete data produces drafts with flags and zero values rather than crashing.