AI agents for business, each built from one sentence.
An AI agent takes a job, uses your tools to do it and reports back. The agents here run in Arclyx today. Each began as one sentence, such as sort the inbox, find tenders worth a bid or reconcile the bank. Every page shows what the agent reads, what it posts and how to build your own.
What is an AI agent?
An AI agent is software you give a goal instead of steps. It decides what to do next, uses tools such as your email, CRM, documents or the web to do it, and reports what it did. A chatbot answers; a fixed workflow repeats the same steps; an agent does the work and adapts to what it finds.
AI agents vs chatbots and automation.
All three save time. They differ in what you give them and what you get back.
| Chatbot | Workflow automation | AI agent | |
|---|---|---|---|
| Starts from | A question you type | A trigger and steps you draw | A job in one sentence |
| Works with your tools | Only what you paste in | Yes, in fixed steps | Yes, choosing the next step |
| When the input changes | Answers the new question | Follows the same path, or stops | Adapts to what it finds |
| What you get | A reply in a chat | The same steps, every time | The work done, and a report |
AI agents for sales teams
Find who to call, walk into every meeting prepared, and catch the tenders worth a bid.
“Find up to 10 B2B software companies headquartered in Canada with 50 to 200 employees and at least one open sales role in Canada, name their head of sales, and post the list with a source for every fact to #prospects.”
AI agents for marketing and product teams
Know what competitors changed, what customers keep saying, and get a cited first draft.
“Every Monday at 8, check Competitor A, B and C for pricing, product, news and hiring changes since last week, compare them with our competitor notes, and post a cited digest to #competitive.”
AI agents for finance teams
Reconcile the bank, report the month's KPIs, and pull month-end into one message.
“On the 2nd of each month, run our month-end: have Bank reconciler reconcile the bank and KPI reporter report the KPIs, build a 30-day cash outlook from our tables, check every number, and post one summary to #finance.”
AI agents for operations teams
Start the day knowing which emails need you.
“Every weekday at 17:00, sort my unread Gmail from today into needs a reply, FYI or can wait, and post it to #my-inbox.”
What goes into an AI agent?
One sentence holds all of it. This is the one that built Inbox triage.
Every weekday at 8, read my unread Gmail from the last 24 hours, sort each email into needs a reply, FYI or can wait, and send me one summary in #my-inbox with a next step for each reply.
- A schedule, an @mention or a webhook starts it.
- The tools it may read, and nothing else.
- The judgement you'd otherwise do yourself.
- One place it may write: a channel, or an email to you.
- What lands, and what you do with it.
How do you build an AI agent?
In Arclyx, the same way every agent here was built: describe it, check it, let it run.
Describe the job
One sentence: what to do, when, and where the result goes.
Arclyx drafts the agent
It writes the instructions, picks the tools and asks about anything it can't work out.
Test it on your real tools
One run, round by round, before it posts anything you'd rely on.
Let it run
On a schedule, when you @mention it, or when another system calls it.
What every agent here has in common.
Reads more than it writes
Each agent reads the tools it needs and writes to one place: a Slack channel, or your own inbox.
Every run is recorded
What it searched, read and decided, step by step, for anyone to check.
A ceiling on every run
A run that would pass its spend limit stops and tells you.
Agents hand work to agents
One agent can ask another for its part and check the result before using it.
AI agents, answered.
What are some examples of AI agents?
Agents that sort your unread email, brief you before external meetings, find public tenders worth a bid, watch what competitors changed, turn customer feedback into themes, draft cited blog posts, reconcile the bank and report the month's KPIs. Each one on this page runs in Arclyx and links to how it works.
What can AI agents do for a business?
Take on the recurring work that needs reading and judgement, not just copying: checking sources, comparing against what your team knows, and writing up what changed. They work from your own tools and post the result where you already look: a Slack channel, or an email to you.
How is an AI agent different from a chatbot?
A chatbot answers what you ask in a chat window. An agent is given a job, uses your tools to do it on its own, on a schedule or when asked, and reports back with what it did and where each fact came from.
How is an AI agent different from workflow automation?
Automation follows steps you draw in advance, the same way every time. An agent decides its next step from what it finds, such as opening a page only when a snippet isn't enough, while staying inside the tools and limits you give it.
How do you build an AI agent?
In Arclyx you describe the job in one sentence. Arclyx drafts the instructions, connects the tools, asks about anything unclear and runs a test on your real data. Then you choose when it runs. You don't write code.
Are AI agents safe to use with company data?
They are when access is narrow and visible. Each agent here reads only the tools it lists, writes to one place, can email no one but you, records every run and stops at a spend ceiling. Read-only agents can't change your inbox, books or CRM.
Build your own agent.
One sentence in, a working agent out. Join the beta and we'll set you up.