A tour of the agent detail page Guide

Every agent you create has its own detail page — the single place where you shape what it does, review what it has produced, and adjust how it keeps you posted. This tour walks through each area so you always know where to look.

The Build tab

Build is where most of the work happens. You describe what you want the agent to do in plain English, and the AI builder — named Stan — sets it up for you, wiring in the right tools, skills, knowledge, and notifications as it goes.

Because Build handles the heavy lifting, you rarely need to configure things by hand. If your agent is missing a capability or following the wrong approach, the fastest fix is usually to tell the builder what you want changed — for example, "let it send WhatsApp messages" or "have it write in our brand voice."

Tip: When something isn't working the way you expect, describe the outcome you want on the Build tab rather than hunting for a setting. The builder translates your request into the right configuration.

The Configure nav

The left-hand navigation groups five tabs under Configure. Each one gives you a window into a different part of what makes your agent work. You can view all of them, and you can directly manage Knowledge and Secrets yourself.

Tools

The Tools tab shows everything your agent can actually do. It lists the built-in tools available to it — things like emailing you, saving results, searching the web, working with files, and generating images — alongside any connected integrations such as Gmail, Slack, or Shopify.

Integrations are grouped so you can see which connected account each one uses at a glance. Built-in tools are always available; integrations require you to connect an account first before the agent can use them. If an integration appears but isn't connected yet, the agent can't use it until you finish connecting it.

Skills

Skills are packaged expertise — proven, step-by-step playbooks for a specific job, such as writing an SEO blog post or monitoring a brand. The Skills tab shows which skills your agent has installed.

Installing a skill also grants the extra tools that skill needs, so your agent gets both the method and the means in one step. This is why an agent sometimes has tools you don't remember adding — a skill brought them along. Skills are added mainly through the AI builder, so if your agent isn't following a proven method, ask the builder to install the matching skill.

Files

The Files tab holds two things. First, the agent's own "mind" — its persona and memory files, which capture who it is, its purpose, what it knows about you, and what it has learned over time. Second, every output file it has produced, such as PDFs, images, and reports.

You can switch between agent files and output files, then open, preview, or download any of them. Files are read-only for you; to change one, ask the builder. If you don't see any agent files yet, they're created on the agent's first run. Outputs also appear on the Deliveries tab.

Secrets

Secrets is a secure vault for the API keys and tokens an agent needs to call outside services — a webhook token, a third-party API key, and the like. Values are encrypted and hidden after saving, and the agent references them by name rather than ever seeing the raw value.

You can manage secrets directly here:

1. Add a secret with a name, value, and an optional description.
2. Edit a secret's value or description whenever it changes.
3. Delete any secret the agent no longer needs.

The tab shows at a glance which secrets are set and which are still empty. If an agent fails to reach an external service, an empty or missing secret is a common cause — check that the expected name has a value.

Knowledge

The Knowledge base is everything your agent reads on every run — product docs, brand guidelines, price lists, FAQs — plus facts it picks up on its own as it works. Good knowledge grounds the agent in your specifics so its output stays accurate and on-brand.

You can manage knowledge directly:

• Upload files (PDF, Word, text, CSV, or Markdown) or drag and drop them in.
• Add a text note with a title and body.
• Review what the agent already knows and remove anything outdated, like a superseded price list.

Note: Knowledge is what the agent reads every run, while Files are its own mind and the outputs it produces. If your agent is giving generic answers, it may simply be missing key knowledge.

The Settings tab

Settings controls how the agent keeps you posted on its work — for example, emailing you a summary, pinging you on a chat channel, or staying quiet. Like the rest of the agent, you can set this up in plain English through the builder, so the agent reaches you where you actually are.

Confirm and roll back changes

After you adjust an agent, you can run a test to confirm it behaves the way you intend before it goes to work for real. Version history lets you see what changed and roll back if a change didn't land the way you hoped — so you can experiment with confidence.