Use cases

A tech blogger connected Paperless-ngx 3.0 to a language model and tried its Suggest button, which proposes tags, a document type, a correspondent, and a title. Nothing changes until a person accepts a suggestion.

An automation builder runs a scheduled agent that reads leads in Airtable, checks Gmail for replies, and saves follow-up drafts. The drafts go to Slack, where the builder approves or rejects each one.

A team built a Claude Cowork skill that gathers the day's calendar, recent messages, and open tasks into one brief each morning. The brief is a summary: a person reads it, and the post describes no step where anything is sent.

A consultant uses ChatGPT or Claude to draft a three-step series of follow-up emails for late invoices, saved as templates. The consultant adds the client details and sends each email personally from their own mailbox.

A freelance paid media consultant uses Claude to draft the executive summary, campaign commentary, and anomaly notes of monthly client reports. The consultant checks the data first, edits the draft, and lays out the final report.

A New Zealand Google Partner agency built an audit tool in which Claude writes a 17-section report from Google Ads API data and the agency's own best-practice file. Staff review the output and can edit every section before the audit is exported as a client PDF.

A marketing lead at a startup runs one Claude Code command each Sunday evening to prepare the next week's meeting documents. The lead reviews every line, and nothing reaches a live document until the lead signs off.

A marketer keeps procedures, checklists, and house rules in one Obsidian folder that Claude Code works from. Finished work waits on a board for accept, ask, or deny, and paid API calls stop at an approval note first.

A hosting provider describes paperless-gpt, a service that proposes titles, tags, correspondents, document types, and dates for scanned documents in Paperless-ngx. Documents with the review tag wait in a screen where a person approves each change.

A performance marketing agency has an AI model write plain-language summaries of the weekly changes in each client's campaigns. Each summary is saved as a draft, and the account manager reviews, adjusts, and sends it.

A professional writes a rules file that tells Claude Cowork how to sort the morning inbox and draft replies, then reads a one-page summary. The Gmail connector can only create drafts, so the author reads and sends every reply.

A sales team uses an AI agent to draft the follow-up email after each sales call, based on the call summary. The rep edits every draft and sends it personally.

A company has an AI agent check competitor pages and web mentions every morning and write one brief per competitor. The brief is internal reading material for the team, and the agent sends nothing to competitors.

A technology executive has Claude Cowork prepare a briefing of urgent email, the day's calendar, and a short action list every morning. The author reads it, and the post shows no step where anything is sent.

A marketing lead at a small company has AI agents read six marketing systems in one pass and return one page on what changed. The agents draft and analyze, and a person approves anything that goes to customers.

A one-person business built a Claude Code skill that reads receipts from its Fiken inbox and proposes an account for each one from past bookings. The owner confirms every proposal before anything is booked.

A small web-design agency used an n8n workflow to classify the messages in its Outlook inbox and draft replies for the harder ones. The owner approves each of those drafts with one click, and the workflow logs its actions to a Google Sheet.

A company co-founder set up a weekly Claude Cowork task that turns recent closed-won deals into draft outreach for similar companies. The drafts are posted to Slack where the team reviews them, and sending is not automated.

An agency with 11 clients has AI agents draft weekly Google Analytics 4 reports on Wednesday night. Each account manager reviews and edits the drafts on Thursday before they go to the clients.

An n8n workflow template reads each new Gmail message with a Gemini agent, labels it, and saves a reply draft where one is needed. Each draft waits in Gmail, with the message marked unread, for a person to review.

easybill's help article describes a connector that lets an AI assistant create invoices, quotes, and credit notes in an easybill account. Each document starts as a draft, and finalizing, sending, and recording payments wait for the user's confirmation.

Google documents an assistant inside Google Ads that answers questions about campaign performance, ad disapprovals, and billing, and suggests changes. The advertiser approves a change before it is applied.

SimplyBills describes how a freelancer connects their own ChatGPT or Claude to the invoicing tool so the assistant finds overdue invoices and drafts the reminder. Nothing is sent until the user confirms, and sending through the assistant is limited to the Max plan.

The Trade Desk describes conversational Koa agents in its Kokai platform that help build a campaign from an uploaded media plan, flag setup problems, and troubleshoot campaigns that are not spending. The feature is in closed beta, and the buyer decides what to apply.