Prompting and Input

You spend most of your time in Polytoken at the prompt, writing messages, referencing files and tools, and scrolling through the conversation.

Writing a prompt

The prompt is a multi-line text area. Type into it and press Enter to send. To add a line break without sending, press Shift+Enter. The exact modifier depends on your terminal.

Prompts you send while Polytoken is working go into a queue and arrive on its next turn.

Referencing things with `@`

Type @ anywhere in the prompt to open the reference picker. As you type after the @, Polytoken fuzzy-matches candidates and shows them in a bar above the prompt. Matching is case-insensitive until you type an uppercase letter, at which point it becomes case-sensitive.

The picker offers four categories:

You want to reference	Type	Short form
A file in your project	`@path/to/file`
A skill	`@skill:name`	`@s:name`
A subagent	`@subagent:name`	`@a:name`
A model	`@model:name`	`@m:name`

File references are limited to your project. By default, Polytoken leaves out hidden files, the project’s Polytoken directory, and anything covered by .gitignore, .claudeignore, or .polytokenignore.

Two extra controls appear while the picker is open, and the bar shows them when they apply:

Shift+Tab toggles whether the file picker respects those ignore rules, letting you select an ignored file when you need one.
[ and ] adjust the reasoning level when you have a model reference selected.

Running commands with `/`

Type / as the first character of an empty prompt to open the command palette. A / later in your text is just text, unlike @, which works anywhere.

Some commands run immediately, some open a second menu to pick a value such as a model or a facet, and some take text you type after them:

/model switches the active model; /facet switches the active facet.
/compact and /clear manage your context. See Sessions.
/rewind returns the conversation to an earlier point. See Sessions.
/permissions changes how tool approvals are handled. See Permissions.
/todo and /jobs show your todos and running jobs. See Managing Work.

The full list is in the Command Reference.

The typeahead bar

The reference picker and the command palette share the same bar above the prompt and the same keys:

Up / Down move through the candidates.
Tab or Enter accepts the highlighted one.
Esc dismisses the bar.

The bar labels the keys that apply at each moment, including the file-ignore and reasoning controls when they are relevant.

A facet decides what your prompt does. The same prompt (“fix the login bug”) leads to different behavior depending on which facet is active. A facet sets how Polytoken frames the model, which tools the model can use, and the accent color of the session.

Polytoken ships two facets. Plan is a read-only thinking facet: the model investigates, asks questions, and writes a plan, but it cannot edit files or run commands, because the plan facet does not provide the edit and shell tools. Execute is the working facet: the model has those tools and uses them.

Switch facets with /facet plan or /facet execute. The sidebar updates to show the new facet and its color, and the change takes effect on your next prompt.

Facets and models are independent: switching one does not switch the other. You can plan and execute on different models, or hold one model across both. You can also write your own facets; see Facets in Harness Engineering.

Plan mode

Plan mode is the plan facet, seen from the moment the plan is ready. The model cannot move into execution on its own, so when the plan is done it hands the plan to you and waits for your decision.

At the handoff you get three choices:

Implement in a new context. Switch to execute and start fresh. Polytoken clears the planning conversation and carries the plan forward.
Implement in the current context. Switch to execute and keep the whole conversation, plan and all.
Cancel. Stay in plan. The model can revise the plan and hand off again.

The first choice clears the conversation with the same operation as /clear; see Sessions. The plan survives the clear.

Moving through the conversation

The conversation above the prompt is a stack of cards. A card is one selectable piece of the conversation: your prompt, a reply, a tool call, a notice. Most cards start collapsed. A tool card expands to show its result; a compaction marker expands to show what it summarized. Your prompts and Polytoken’s replies always show in full.

With the mouse

Scroll to move through the conversation. Scrolling does not interrupt what you are typing.
Single-click a card to expand or collapse it without leaving the prompt.
Double-click a card to step into the conversation with that card selected.
Click and drag to select text within a card.
Double-click a link to copy it.

With the keyboard

Press Ctrl+Up to step into the conversation. From there:

Up / Down move between cards; Shift+Up / Shift+Down page through them.
Left / Right collapse and expand the selected card.
Ctrl+Y copies the selected card’s content.
Ctrl+R opens the rewind view for the selected prompt.
Esc returns to the prompt. Typing does too; the first character you press lands in the prompt.

These keys are the defaults and can be changed in your configuration.