Find and cancel forgotten subscriptions from card statements
Upload three months of statements and get every recurring charge grouped by merchant, with the ones you forgot about flagged and step-by-step cancellation links.
What people are really getting done with AI agents
21 documented use cases across 9 agents. Each one names the agent, links the source and includes the prompt, so you can run it yourself.
/
21 use cases
This banner spot is open. Reach people deciding which agent to use next.
See sponsorship optionsDaily inbox triage with drafted replies
Every morning a local agent sorts new email into reply, read later and ignore, drafts replies for the first group, and sends you a short briefing in chat.
Compile a Monday operations digest from Slack, Linear and email
Every Monday the agent reads last week's channels, tickets and threads and writes one page: what shipped, what is blocked, and which decisions are still waiting on someone.
Turn a folder of receipt photos into a monthly expense report
A self-hosted agent watches a folder, reads each receipt photo, categorizes the expense and appends it to a spreadsheet you can hand to your accountant.
Hunt down and fix flaky tests in CI
The coding agent reruns the suite to identify flaky tests, traces each one to a race, shared state or timing assumption, and opens a fix with an explanation.
Monitor competitor pricing pages and summarize changes
A self-hosted agent snapshots five competitor pricing pages every morning and messages you only when a plan, price or limit actually changes.
Generate thumbnail concepts and title variants from a video script
The agent checks what is currently getting attention on the topic, then proposes title options and thumbnail concepts with generated mockups to test.
Rough market sizing with every number sourced
The agent builds a top-down and a bottom-up estimate, shows the formula for each, and links the source behind every input so you can check it.
Enrich an inbound lead list and score fit
Upload a raw list of signups; the agent looks up each company, adds size, industry and tech signals, and scores fit against your ideal customer profile.
Repurpose one podcast episode into a week of posts
From a single transcript the agent drafts a LinkedIn post, an X thread, three short-clip timestamps and a newsletter section, all in the host's voice.
Map 30 research papers into a literature review
Deep research reads the papers, groups them by method and finding, and produces a literature map with agreements, contradictions and open gaps.
Shortlist vendors and build a side-by-side comparison
Give the agent your requirements and budget; it researches vendors, reads pricing and docs pages, and returns a comparison table with sources for every cell.
Plan a multi-city trip and keep a living itinerary
The agent compares train and flight options between cities, proposes a day-by-day plan within budget, and keeps one document updated as bookings come in.
Upgrade a framework across a monorepo, one package at a time
The agent reads the migration guide, upgrades packages in dependency order, runs tests after each step, and leaves a reviewable commit per package.
Prepare a retention-call script to lower an internet bill
The agent researches current promo prices from competitors in your area and writes a short call script with the offers to cite and when to ask for the retention team.
Book appointments by filling web forms for you
The browser agent finds open slots across clinics or services, compares them against your calendar, and fills the booking form up to the final confirmation.
Convert meeting transcripts into assigned tickets
Drop in a transcript and the agent pulls out every action item, matches owners to your team list, and creates draft tickets with due dates for you to approve.
Turn top error-tracker issues into fix pull requests
Each week the agent takes the most frequent production errors, reproduces them, and opens pull requests with a fix and a regression test.
Curate a weekly niche newsletter from 40 sources
The agent reads a fixed list of blogs, changelogs and forums every Friday, picks the ten most relevant items, and drafts the issue with one-line takes.
Draft an incident postmortem from logs and the chat timeline
Feed in the incident channel export and key log excerpts; the agent rebuilds the timeline, separates cause from symptoms, and drafts a blameless postmortem.
Find live conversations worth joining on X
Each morning the agent finds threads where people are asking about problems your product solves and suggests a helpful, non-promotional reply for each.