Overview and Key Capabilities
Agent Use Cases is a continuously updated directory of real AI agent use cases. Every day it scans the open web, X, Reddit, and Hacker News for accounts of people actually delegating work to agents, then rewrites them into use cases you can copy.
Unlike most AI tool directories, it does not list products. It lists finished, concrete tasks: planning a college-tour road trip with Claude Code, pasting itemized bills into ChatGPT to flag fees and draft a retention call script, or having Copilot add docs and validation to a repo and repair its own CI failure.
Core Features
- Daily scanning: the site ingests new cases every day and surfaces what was added today and this week.
- Two-way filtering: browse by agent (Grok Bot, Claude Code, Claude, OpenClaw, Codex, n8n, ChatGPT, ChatGPT Agent, Claude Cowork, Gemini and more, 28 in total) and by topic (Money, Bills and Benefits; Life Admin and Home; Shopping and Errands; Travel and Bookings; Work and Operations; Sales, Marketing and Support; Research and Knowledge; Content and Creative; Software and DevOps; Games and Simulations; Security, Science and Hardware).
- Case detail: each entry shows the scenario, the agent used, the source link and date, plus the prompt where the source published one.
- Daily 5 digest: an email with five use cases every morning, and an open submission form for new cases.
Scale
The directory currently holds roughly 2,268 use cases, led by Software and DevOps at 431, Work and Operations at 412, Life Admin and Home at 302, and Sales, Marketing and Support at 284.
Who It Is For
- Product discovery: see where real demand clusters before deciding what to build.
- Internal adoption: concrete examples persuade skeptical colleagues better than explanations of capability.
- Personal leverage: filter by the chores you repeat most, such as expenses, bookings, or research, and copy what already works.