CtrlOps
|Docs
Product ModulesAI Terminal

AI SSH Terminal

CtrlOps AI SSH Terminal is a full Linux terminal with a built-in AI terminal assistant. Ask questions in plain English, get the right Linux commands, review before execution, and let AI explain every output. It turns natural language into bash - instantly.

The AI Terminal is the same SSH terminal you'd use anywhere, with as many tabs as you need and two side panels you can pop open: Ask AI, which turns plain English into commands and explains output, and a Scripts library where you keep the commands you run all the time. Type a command yourself, ask the AI to figure one out, or run a saved script in one click. For an overview of the AI-powered terminal before you set it up, see the feature page.

What is an AI terminal?

An AI terminal is a command-line interface enhanced with artificial intelligence. Instead of memorizing Linux commands, you describe what you need in plain English - the AI generates the correct command, explains what it does, and executes it with your approval.

Traditional SSH terminals require you to know exact syntax. An AI terminal assistant bridges that gap: it understands context, reads server output, chains multiple commands, and provides human-readable summaries of results.

CtrlOps takes this further by combining a full SSH session with an AI assistant panel side-by-side. You can type commands manually when you know them, or ask the AI when you don't - all in one window, connected to your live Linux server. No server-side agent required.

What you can do with the AI SSH terminal

  • Type any Linux command and see the output, just like a regular SSH terminal.
  • Open as many terminals on one server as you need, each in its own tab.
  • Read output as a plain stream, or as one block per command you can copy, filter and re-run.
  • Get command suggestions as you type, with a one-line description of what each one does.
  • Ask the AI plain-English questions ("what services are running?", "why is the disk full?").
  • Have the AI run commands for you, with your approval before each one.
  • Connect a ChatGPT or OpenRouter account in one click, or bring your own OpenAI, Anthropic, or Gemini key, and switch between them mid-conversation.
  • Save commonly-used commands as Scripts and re-run them with one click.
  • Let the AI search the web for current docs, error messages, and package versions.
  • Use AI as your Linux terminal assistant - debug errors, optimize configs, and manage services without memorizing commands.
  • Works with any Linux server you connect via SSH - Ubuntu, Debian, CentOS, AlmaLinux, and more.

Open the AI SSH terminal

Open app → Click your server → Click the AI Terminal tab

You'll land in a terminal taking up most of the screen. Along the top is a strip of terminal tabs with a + at the end, and to the right of that the controls for the tab you're in:

ControlWhat it does
Terminal tabs and +Each tab is its own shell on this server. + opens another
Ask AIOpen the AI panel that turns plain English into commands
ScriptsOpen your saved script library
View toggle (two icons)Switch between Terminal view and Block view
Suggestions (toggle + gear)Turn command suggestions on or off, and configure them

Both the Ask AI and Scripts panels are collapsed by default.

AI Terminal in the left sidebar opens the terminal. The tab strip and header controls run along the top.

Work in more than one terminal

One shell is rarely enough. Tail a log in one tab, run a build in another, and keep a third free for whatever comes up.

Click + at the end of the tab strip to open another terminal on the same server. Each tab is a separate shell with its own working directory, so a cd in one doesn't move the others, and a long-running command in one doesn't block the rest.

Tabs name themselves after the command you run, so a tab where you ran pm2 becomes pm2 rather than Task 2. To name one yourself, edit the tab's name directly, which is worth doing when you're keeping a tab open for a specific job.

These are several terminals on one server, not one command sent to several servers. CtrlOps never runs a command across your fleet: to work on another server, open it in its own server tab.

Use the terminal directly

Click anywhere in the dark area and start typing. It works exactly like an SSH session:

  • Hit Enter to run a command.
  • Tab completes what you're typing. Exactly what it does is yours to set, see Suggestions below.
  • Up arrow to repeat the previous command.
  • Scroll wheel to look at history (5,000 line buffer).
  • Ctrl+C / Cmd+C copies selected text.
  • Ctrl+F / Cmd+F opens a search overlay.
  • Right-click for Copy, Paste, Clear Buffer, Reset Session.

If your connection drops, you'll see a "Disconnected" alert in the bottom-left with a Reconnect button.

Terminal view and Block view

The two icon buttons in the header switch how output is drawn. The shell underneath is identical either way, so you can flip between them whenever you like.

ViewWhat you get
Terminal viewPlain shell output, exactly as the server prints it
Block viewKeeps each command with its own output in one block, so you can copy, filter or re-run it later

Terminal view is the right choice when output is meant to be read as one continuous stream, a tail -f for instance, or when a full-screen program like htop or vim takes over the window. Block view earns its place afterwards, when you're working through a list of commands and want to go back to what one of them printed without scrolling through everything that came after it.

The two icon buttons in the header switch between Terminal view and Block view.

Command suggestions as you type

Start typing and CtrlOps suggests the rest. A small dropdown appears above the cursor with matching commands, subcommands, and flags - each with a one-line description of what it does. Type pm2 l and you'll see list, logs, logrotate, login, and link side by side with what each one actually does, so you no longer have to remember the full syntax for the commands you only run once a month.

KeyWhat it does
↑ / ↓Move through the suggestions
EnterAccept the highlighted suggestion
TabDepends on your Tab key setting, see below
EscDismiss the dropdown and keep typing

The suggestions narrow as you type, so you can keep going and ignore them entirely if you already know the command. This is plain autocomplete and it costs you nothing - it doesn't call the AI and doesn't need an API key.

Configure suggestions

The Suggestions toggle in the header turns the whole thing off, which is what you want if you find the dropdown distracting. The gear beside it opens the settings.

The gear beside the Suggestions toggle opens Tab key behavior and the three checkboxes.

Tab key behavior decides what Tab does, and it's the setting worth getting right first:

OptionWhat Tab does
Open completions menuBring up the list of matching commands
Accept autosuggestionTake the greyed-out completion sitting at the end of your line

Three checkboxes cover the rest:

SettingWhat it controls
Open completions menu as you typeShow the dropdown automatically. Turn it off to only see it when you ask for it
AutosuggestionsThe inline greyed-out completion at the end of the line
Show autosuggestion keybinding hintDisplay which key accepts that completion

If Tab isn't behaving the way your usual shell does, this is why. Set Tab key behavior to match the habit you already have rather than learning a new one.

Add an AI key (one-time setup)

The AI Assistant needs one AI provider connected before it can answer anything. There are three ways to do that, and two of them need no API key at all. All three start on the same screen: in the left sidebar, under TOOLS, click AI Keys.

Your situationUse thisWhat you need
You already have a ChatGPT accountSign in with ChatGPTYour ChatGPT login. No API key.
You don't pay for any AI serviceConnect with OpenRouterA free OpenRouter account. CtrlOps picks a free model for you.
You want a specific provider or model, or already hold a keyAdd credentialAn API key from OpenAI, Anthropic, Google, or a custom endpoint.

The top of the AI Keys page is Quick connect, with the ChatGPT and OpenRouter cards side by side. Below it is API credentials, where your own keys live.

Quick connect sits at the top of the AI Keys page, with API credentials underneath. The OpenRouter card here is already connected and running a free model.

Sign in with ChatGPT

Pick this if you already have a ChatGPT account. CtrlOps signs in through OpenAI, so there is no API key to generate, copy, or paste.

Open the AI Keys page

In the left sidebar, under TOOLS, click AI Keys.

AI Keys sits under TOOLS in the left sidebar.

Click Sign in with ChatGPT

It's the button on the first Quick connect card, labelled ChatGPT account.

Sign in with ChatGPT is the button on the first Quick connect card.

Approve the sign-in in your browser

CtrlOps opens your browser. Sign in to your OpenAI account, approve access, and the browser hands you back to the app.

Check the card

The ChatGPT card now shows your connected account and a sign-out icon, and the model is selectable from the AI Assistant dropdown straight away.

Once connected, the ChatGPT card shows a green dot, a MODEL dropdown and a sign-out icon.

This path runs on the ChatGPT plan you already pay for, so there's no per-token API bill and no key to store or rotate. To disconnect, click the sign-out icon on the ChatGPT card.

Connect with OpenRouter (free)

Pick this if you don't pay for ChatGPT. OpenRouter is one account in front of hundreds of models, and it serves a set of them free. You do not need a paid OpenRouter key: when you connect, CtrlOps picks the best free model available and selects it for you.

Open the AI Keys page

In the left sidebar, under TOOLS, click AI Keys.

AI Keys sits under TOOLS in the left sidebar.

Click Connect with OpenRouter

It's the button on the second Quick connect card, labelled OpenRouter account.

Connect with OpenRouter is the button on the second Quick connect card.

Approve the connection in your browser

Sign in to OpenRouter, or create an account if you don't have one, then authorise CtrlOps. Creating the account is free and doesn't need a card.

Let CtrlOps pick the model

Back in the app, the OpenRouter card shows a MODEL dropdown already set to a free model. Leave it as it is, or open the dropdown to switch to a different one at any time - there's nothing to reconnect.

The OpenRouter card comes back already set to a free model. Open the MODEL dropdown to switch at any time.

Free models are served by third-party providers who may log your prompts and responses, and they're rate-limited. Don't paste secrets, credentials, or confidential server output into a chat backed by a free model.

Free models are ideal for learning your way around the AI Terminal. For real server work - running commands, debugging incidents, touching infrastructure - a high-capability model like Claude Opus, Gemini 2.5 Pro, or GPT-5 earns its cost, because smaller models are more likely to suggest commands that misfire.

Add your own API key

Pick this when you already hold a key from a provider, or you need a specific model that the two Quick connect options don't give you. It works with OpenAI, Anthropic, Google, and any OpenAI-compatible endpoint.

Open the AI Keys page

In the left sidebar, under TOOLS, click AI Keys.

AI Keys sits under TOOLS in the left sidebar.

Click Add credential

The button sits at the top right of the API credentials section. If you have no credentials yet, the same button is in the empty state, under Connect your first AI provider.

Add credential sits at the top right of the API credentials section, below Quick connect.

Fill in the Add AI credential form

FieldWhat to enter
Name (identifier)A label for you, like Prod LLM or Personal Key. Up to 50 characters.
ProviderDefaults to OpenAI. Switch it to Anthropic, Google, or a custom OpenAI-compatible endpoint.
API keyPaste the key from your provider's console. The eye icon reveals it so you can check the paste.
ModelReads Enter credentials to load models until the key is accepted, then fills with your provider's model list.
Base URL (custom endpoints only)The endpoint URL including the version path, e.g. https://api.your-provider.com/v1 - not just the host.
The Add AI credential form. The Model list fills in once the key is accepted.

Click Test connection

CtrlOps makes a tiny test call. You'll see a success toast or a specific error.

With the form filled in, Test connection sits between Cancel and Save credential.

Click Save credential

The key is stored locally on your computer and encrypted at rest, never on a CtrlOps server. Saving runs the test again automatically; the credential saves either way, marked as error status if the test failed.

Save credential sits at the bottom right of the form, next to Test connection.

You can add as many credentials as you want. The AI panel lets you switch between them, and between the Quick connect providers, mid-conversation.

Open the AI Assistant panel

In the AI Terminal header → Click Ask AI

A 450px panel slides in from the right with the chat history (empty at first), an input box at the bottom, and a green circular send button.

If you haven't connected an AI provider yet, you'll see "AI Not Configured" with a button that takes you to the AI Keys page.

Ask the AI a question

The AI terminal assistant understands natural language. Describe your problem or goal, and it generates the right Linux commands for your server.

Pick which key to use (optional)

At the bottom-left of the AI panel, a dropdown shows the currently selected key. Click it to switch to a different provider or model. The next message you send uses the new key.

The key dropdown at the bottom-left of the AI panel switches provider or model for the next message.

Type your question

Use plain English in the input box at the bottom. Examples:

  • "What services are running on this server?"
  • "Why is /var/log so big?"
  • "Set up a daily cron job that runs my-script.sh at 2am"
Type a plain-English question into the input box at the bottom of the AI panel.

Send it

Press Enter, or click the green send arrow

(Use Shift+Enter for a newline if you want a multi-line prompt.)

Press Enter or click the green send arrow to send the question.

Approve any commands the AI wants to run

If the AI suggests a command, it appears as a card with the command in a code block, an explanation, and two buttons:

ButtonWhat it does
Run (green play icon)Runs the command in the terminal. Output is captured and sent back to the AI
Cancel (red X)Rejects the command. The AI may suggest something else

The AI may chain several commands in a row, asking for approval each time, until it has enough information to give you an answer.

Each suggested command arrives as a card with an explanation. Read it, then click Run or Cancel.

Always read the command before clicking Run. The AI is helpful but it's not infallible. If a suggested command would touch important files or stop a running service, double-check first.

Tweak how the AI behaves

The AI panel has a few controls in the footer.

ControlWhat it does
Max AI reasoning steps (number field, 1-200)Caps how many back-and-forth steps the AI takes per question. Default is fine for most.
Auto-Run (switch)Skip the manual approval step and run AI commands automatically. Convenient but riskier.
Web Search (pill + toggle)Let the AI look things up online for current docs, error messages, package versions, etc. See Web search.
New (header button)Start a fresh conversation, useful when context gets stale
Export (header button)Save the chat as a file for sharing or notes
Stop (red X, only while running)Interrupt the AI mid-thought
Close (X)Collapse the panel

Leave Auto-Run off until you've used the AI for a while and trust the kind of commands it suggests. With Auto-Run on, every command runs immediately with no approval step.

Go deeper

Tips

When something is broken, paste the exact error into the AI Assistant and ask "why?". The AI is excellent at decoding cryptic error messages.

When the AI keeps suggesting wrong commands, click New in the panel header to start a fresh conversation. Long contexts can drift.

Keep two AI keys configured: one fast and cheap (Haiku, GPT-4o-mini, Gemini Flash) for quick lookups, one heavy (Opus, GPT-4o, Gemini Pro) for hard debugging. Switch between them from the dropdown.

The AI SSH terminal works best when you give it context. Instead of "fix nginx", try "nginx is returning 502 after I updated the config - check the error log and suggest a fix." More context = better commands.

Why use an AI-powered Linux terminal?

Managing Linux servers through a traditional SSH terminal means memorizing hundreds of commands, parsing cryptic error messages, and spending time searching documentation. An AI-powered Linux terminal assistant handles this for you:

  • Diagnose server issues in seconds: paste an error, get the fix.
  • No command memorization: describe what you need, the AI writes the command.
  • Approve before execution: every AI-generated command requires your approval (unless you enable Auto-Run).
  • Switch AI providers freely: sign in with ChatGPT or connect OpenRouter in one click, or use OpenAI, Anthropic Claude, Google Gemini, or any OpenAI-compatible provider with your own API key, stored locally with AES-256 encryption.
  • Secure SSH connections: connect to your Linux server via SSH with no server-side agent needed. Manage your credentials and keys from the SSH Key Management module.
  • Go beyond debugging: once you've diagnosed and fixed an issue, deploy updates with one-click deployment without leaving CtrlOps.

Whether you're a solo developer managing multiple client servers or a startup CTO without a dedicated DevOps team, the AI SSH terminal turns server management from a specialized skill into a conversation. It is also one of the key differentiators that makes CtrlOps the best Termius alternative for AI-powered server management.

Troubleshooting

No AI provider is connected yet. Open AI Keys from the left sidebar, under TOOLS. Under Quick connect, click Sign in with ChatGPT if you have a ChatGPT account, or Connect with OpenRouter to use a free model with no API key at all. To bring your own provider key instead, click Add credential in the API credentials section. See "Add an AI key" above.

The API key is wrong, expired, or doesn't have permission. Open the AI Keys page, click Edit on the credential, paste a fresh key, and click Test connection before saving.

The provider's API isn't returning a model list. Most often this means the API key is wrong, or your provider needs a custom Base URL. Double-check both, then click Test connection again.

The terminal session may have disconnected. Look for a "Disconnected" alert at the bottom-left, and click Reconnect. Then ask the AI to retry.

The session may be hung. Right-click and choose Reset Session, then Reconnect if needed. Your scrollback is preserved.

The conversation hit the Max AI reasoning steps cap before finishing. Increase the cap in the panel footer (default 200 is plenty for most). For very complex tasks, break them into smaller questions instead.

Tab is configurable now. Open the gear next to the Suggestions toggle in the terminal header and check Tab key behavior. Open completions menu brings up the list of matching commands; Accept autosuggestion takes the greyed-out completion at the end of the line. Pick whichever matches the shell you're used to.

Switch to a cheaper model from the dropdown for everyday questions, and reserve the heavy models for hard problems. Auto-Run can also rack up costs by encouraging long chains of commands, leave it off when you're being mindful of usage.