getsales

GetSales - CLI-First Operating Layer

This skill is the default operating layer for direct GetSales API work.

Start with the operational guide:

• PLAYBOOK.md - the human workflow for thinking through GetSales operations
• SKILL.md - safety rules and the CLI entry points

Use this skill when you need to:

• inspect contact, flow, sender profile, inbox, and reply statistics;
• read replies and classify lead responses;
• export or prepare lead data;
• work with lists, sender profiles, and automations;
• prepare write payloads through a CLI workflow.

Source of Truth

Before ambiguous or write-sensitive operations, check current GetSales documentation:

• OpenAPI index: https://api.getsales.io/api/openapi
• Quick start: https://api.getsales.io/
• Import guide: https://help.getsales.io/en/articles/10217814-uploading-and-importing-leads-into-getsales

Local notes and old scripts are operational memory, not the source of truth.

Core Rules

1. Run direct GetSales operations through: python3 getsales/getsales.py <command> [args]
2. GETSALES_API_KEY must come from the environment.
3. Before any write/update/delete, show the payload and wait for human approval.
4. Use --dry-run first for write commands.
5. If an API filter is known to be unstable, do not rely on it without verification.

Known Gotchas

• /leads/api/leads/search with flow_uuid can fail server-side. For flow-level work, prefer message or lead-envelope aggregation when possible.
• The message/inbox firehose is noisy. For broad reads, narrow by lead_uuid, conversation_uuid, sender_profile_uuid, or type first.
• Validate import payloads or CSV files before sending anything to the API.

Skill Folder

• Playbook: getsales/PLAYBOOK.md
• CLI: getsales/getsales.py
• Helper scripts: getsales/scripts/ if you add your own

Basic Commands

# Documentation
python3 getsales/getsales.py docs

# Lists / flows / sender profiles
python3 getsales/getsales.py lists
python3 getsales/getsales.py flows
python3 getsales/getsales.py sender-profiles

# Contacts
python3 getsales/getsales.py contacts-search --filter-json '{"company_name":"Acme"}'
python3 getsales/getsales.py contact-get <lead_uuid>
python3 getsales/getsales.py contacts-export --filter-json '{"list_uuid":"..."}' -o leads.csv

# Inbox / stats
python3 getsales/getsales.py inbox --type inbox --limit 50
python3 getsales/getsales.py stats-messages --type inbox --max-pages 10
python3 getsales/getsales.py stats-contacts --filter-json '{"list_uuid":"..."}'
python3 getsales/getsales.py replies --limit 50
python3 getsales/getsales.py reply-intel --limit 50
python3 getsales/getsales.py replies-export -o replies.csv --max-pages 10
python3 getsales/getsales.py stats-replies --max-pages 10
python3 getsales/getsales.py stats-flows --max-pages 10
python3 getsales/getsales.py stats-senders --max-pages 10

# Write with preview first
python3 getsales/getsales.py contact-upsert <list_uuid> lead.json --dry-run
python3 getsales/getsales.py flow-add-new <flow_uuid> <list_uuid> lead.json --dry-run
python3 getsales/getsales.py flow-start <flow_uuid> --dry-run

reply-intel and stats-replies use heuristic intent classification (not_interested, wrong_person, redirect, timing_later, meeting_signal, existing_customer, acknowledgement, auto_reply, and others). Always review important edge cases manually.