Core Concepts

There are four words you need to know to use UpServe well.

1. AI employee

The actor that does the work. Think of it as one member of staff; this guide also calls it an employee for short.

Each employee runs at an AI performance tier you choose. Standard is the default; move an employee up to Ultra (💪) when it needs harder judgment. Which AI model backs each tier is decided by UpServe. See Hire Your First Employee.

When you sign in, you land on Home, and the employees you’ve hired are listed in the employee table, like a staff roster. By default it shows each employee’s name, status, and last run, and the [+] button at the end of the table’s header row adds columns such as AI performance tier or skills and tools.

What is the Manager? One AI manager comes with your account, and it does two things: keeping track of what your employees are working on and telling you about it, and guiding you through the service. 🔴 The Manager does not create employees. New employees come from the Hire button on screen, and the new hire asks you what the job is themselves.

Each employee has:

• Name and role description: what this employee is for
• Behavior instructions: personality, tone, priorities (written by the employee itself during Work setup, right after hiring)
• Tool list: capabilities it can use
• Employee memory: a private store of know-how learned from past runs, visible only to this employee and shown on the Memory & instructions screen. Not shared with other employees.
• Run conditions (optional): scheduled runs (set times), webhooks (external signals), @mentions (called by an employee), self check-ins, and auto memory cleanup (every N conversation turns or scheduled runs). (see How AI Employees Wake Up)

┌───────────────── One AI employee ──────────────────────┐
│                                                        │
│  🪪 Name · role description                            │
│  📜 Behavior instructions                              │
│                                                        │
│  ┌────────────┐  ┌─────────────┐  ┌──────────────────┐ │
│  │  🔧 Tools  │  │  📦 Skills  │  │  🧠 Memory       │ │
│  │ web search │  │   "Notion"  │  │   feedback.md    │ │
│  │    bash    │  │   "Slack"   │  │   project.md     │ │
│  │     ...    │  │     ...     │  │      ...         │ │
│  └────────────┘  └─────────────┘  └──────────────────┘ │
│                                                        │
│  ⏰ Run conditions ─ scheduled run / webhook / mention │
│                                                        │
└────────────────────────────────────────────────────────┘

You can also reply to a specific earlier message instead of just typing a new one. Every chat bubble has a reply icon below it. On web you need to hover over the bubble to reveal it (on mobile web it’s always faintly visible), while in the app (iOS and Android) it’s always shown. Tap it and a quote chip attaches above the composer, showing that your next message continues from that bubble. It’s useful when you want to correct or redirect a specific part of an earlier answer without any ambiguity about what you mean.

You can also speak instead of typing. Tap the microphone button in the composer and talk; your speech is transcribed and dropped into the message box. On web, you check the text and tap send yourself; in the mobile app, it sends automatically after a brief window to cancel. See the mobile app guide for the details on the app’s behavior.

After an employee finishes a task and delivers its answer, you may sometimes see a next steps card suggesting what to do next. You can pick one of up to three suggestions or wave them all off. See the Advanced section below for details.

2. Tool

The employee’s capability. Think of it as a laptop, the internet, or a phone.

Below is a representative set of built-in tools. For the full list, see the Tools guide.

You can toggle tools on and off in employee settings. There’s no separate tool for this; results from autonomous runs reach you the same way a normal reply does, delivered straight to the chat, plus a push notification if you’re not looking at the app.

3. Skill

A reusable ability you add to an employee. Think of it as a work manual.

Where a tool is a single capability, a skill packages multiple tools, code, and task instructions together so an employee does a particular job well.

Examples:

• “Auto-publish to Instagram” skill → image generation + API call + reporting in one
• “Write to Notion” skill → bundles auth, templates, and formatting know-how

You can install skills from Official skills , under the Skills tab, or build your own (yours alone). Once you’ve connected an external service (email, Google Drive, and the like), a skill covering how to use it well is attached to your employees automatically, whether right after you connect it or when you hire a new employee while it’s already connected. You don’t need to go find and install it yourself.

A few concepts come with skills that help you decide which ones to trust and how well they’re holding up:

• Skill score: A combined score built from run volume, success history, and recent use. It is not a guarantee of success rate or stability. It shows as a colored dot on each skill card: green (healthy), yellow (degraded), or red (frequently failing). New skills with no run data don’t show a dot yet.
• Origin & copies: You can copy and edit another skill, and the result is a derived copy that evolves on its own. The badge on a card shows where a skill comes from, making it easier to trace its history.
• Proposals: When an employee finds a defect while running a skill, it automatically queues an improvement proposal for the skill’s author to review. Authors can accept or dismiss proposals and release a new version.
• Curator: A regular background check scans your library for unused, failing, outdated, or deprecated skills and notifies you of anything that may need attention (at most once a week per person). Skills you’ve authored that have had low success rates for a long time, and private skills an employee created autonomously but hasn’t used in a while, can be automatically marked deprecated or archived without asking first. Nothing is deleted, and you’re notified afterward and can restore it anytime. For other cleanup candidates, the curator only flags them and leaves the decision to you.

For a full walkthrough, see Skills and the Autonomous Skill System.

4. Team

Multiple AI employees collaborating as an organization. Think of it as a department.

For complex jobs, splitting roles is more effective than asking one employee to do everything.

Research team
├── Team lead: Plans, splits the work, gathers results
├── Researcher: Handles all web searches
├── Analyst: Analyzes collected material
└── Writer: Produces the final document

When you create a collaborative team, a team lead is added automatically. The team lead plans incoming work, splits it among team members, and gathers the results. It doesn’t carry out the work itself; the team members (the AI employees in the team) do. You can still talk to any team member directly without going through the team lead. Team members pass work to each other with @mentions and report results back. AI employees can also chat directly with each other. Shared team notes (a key-value notepad the whole team can read and write) keep everyone working from the same set of facts. See the Teams guide for details.

Relationship overview

                              Team
                                │
          ┌─────────────────────┼─────────────────────┐
          │                     │                     │
     Employee A            Employee B            Employee C
          │                     │                     │
          ▼                     ▼                     ▼
 Tools·Skills·Memory   Tools·Skills·Memory   Tools·Skills·Memory

💡 Skill = tools + code + task instructions wrapped as a work manual

Words for work

The words for work point at different things. Taking “news summary” as an example:

Advanced

Employee memory and shared team notes

An employee’s memory is its private space for storing know-how, notes, and project context built up over past runs, shown on the Memory & instructions screen alongside its personality, behavior instructions, and work instructions. No other employee can read it. Think of it as a personal notebook.

When information needs to be shared with other AI employees, use the shared team notes: a key-value notepad (for example, "market_analysis" paired with its value) that every employee in the team can read and update. Writing to the same key automatically bumps a version counter, so you never need to encode dates or “latest” in the key name.

Canvas

A Canvas is a versioned document (report, code, plan, or any structured output) that appears directly in the chat view. You can read and request edits without ever downloading a file.

Public version

You can share an employee via an external link or embed it in a website. Outside visitors chat with a public version, a copy of the original employee that runs in its own isolated conversation space; the employee’s private memory is never exposed to outside users. See Sharing an AI Employee Externally.

Run conditions

A run condition is the signal that wakes an employee and starts a run. Beyond a direct message from you, employees can be woken by a scheduled run (a set time), a webhook (a signal from an external service), an @mention (an employee calling their name), a self check-in (the employee sets its own next check-in interval and, when needed, does real work within what it’s allowed to do), or auto memory cleanup (tidying up accumulated memory after a set number of conversation turns or scheduled runs). On the Free plan, though, if you don’t use the service at all for 7 days (no login, message, or chat view on any employee), your active employees are automatically deactivated, pausing every one of these automatic runs. You’ll get a heads-up notification the day before (day 6), and simply using the service again automatically reactivates them. Employees on paid plans are excluded from this automatic deactivation. See How AI Employees Wake Up for details.

Asking you a question (AskUser)

When an employee needs more information mid-task, it pauses and sends you a question. You can reply in plain text, pick from options, or (for things like logins or 2FA) have the employee hand the desktop over so you can complete the step yourself. For sensitive values like passwords, API keys, or tokens, an employee can instead request them through a secure input prompt without taking over the whole screen, and the value is never shown in the chat or to the AI model. You can choose to save it so the employee can reuse it next time.

Reply (quoting a message)

When you reply to a message, the employee doesn’t receive the full original. It receives a short excerpt (up to 240 characters) that the server generates and attaches at save time, sent along with your next message. If the conversation has grown long enough that the quoted message has since scrolled out of the employee’s recent context window, the employee re-fetches up to 2,000 characters of the original text instead of relying on the excerpt alone. The reply attachment clears once you send the message or switch to a different employee or team.

User approvals

Some actions require your approval before they run. When an employee is about to do something sensitive or hard to undo, it pauses and asks you to confirm. See the Approvals guide for details.

Next steps card

When an employee hands you a result (a file, a Canvas document, an image, and so on) and its reply doesn’t already point to what to do next, you may see one more card offering up to three things you could do next. You can pick several at once, or wave them all off with the item at the bottom of the list. It doesn’t appear during automatic runs with no one watching (scheduled runs, self check-ins, etc.) or in conversations shared outside your account.

Voice input

Tap the microphone button in the chat composer and speak; your voice is transcribed and dropped into the message box. If nothing was picked up, you’ll get a notification saying so.

Next steps

• Hire Your First Employee: Get hands-on