Working in the lab · Chapter 8 of 14

Copilot

What it can and can't do, approvals, cost.

~3 min read6 sectionssame text as the guide inside the lab
All chapters

Keys are written for macOS, the way the lab prints them — on Windows and Linux read ⌘ as Ctrl, ⌥ as Alt, ⇧ as Shift, ↵ as Enter.Keys are shown for Windows / Linux to match your platform — on a Mac, Ctrl is ⌘ and Alt is ⌥. Every binding: Keyboard shortcuts.

RoRo Copilot is the last-but-one rail icon: an agent that sees your project and its platform and proposes changes — honest about what it did, what it cost, and what it may not do alone.

What it can see

Reads are automatic, each an activity line.

  • Your project — list_files, read_file (live buffers, unsaved edits included), search_files (string or regex), read_output (the dock's log).
  • The machines — list_machines: codename (the id the SDK, API and Run rail use), qubits, list price per shot, status, queue.
  • A price, from the server — quote_run has the platform price a .qasm for a machine and shots and relays it verbatim. A quote holds nothing and submits nothing; the copilot never multiplies numbers.
  • This project's runs — list_runs / read_run: status, cost, the Output pane's credit line (a refund is certain only when the record says approved), the counts.
  • The library — list_library / read_library_item pull an example's README and program in, so "add a QAOA example" starts from verified code.

What it can do

  • Write — write_file / edit_file pause for an approval card: Edit or New file, the reason, a compact +/− diff. Apply writes through the ordinary save path; Decline is an answer the model must live with.
  • A change set — several files in one round become one card headed N files — a row each with +/−, diff and tick — then Apply / Decline all. Apply writes only the ticked rows, in order (Apply all 3, Apply 2 of 3), and the copilot is told file by file what landed; Stop declines every row.
  • Run Python — run_python asks first, then runs the file on the browser kernel — local, free; the output goes to the model, not the dock.
  • Explain — the Output pane's Explain this result / error is the same metered turn.

From the editor

Select code and a small bar appears — or press ⌘. (the current line when nothing is selected): Explain · Fix this · Add comments · Write a test · Ask about this…. Nothing is sent until you click; then an ordinary turn: file and lines quoted in the transcript, credits charged, Stop ends it, any edit still a card.

What it cannot do

No way to submit a run or spend run credits — a quote is a price check; hardware stays behind the Run rail's confirm. Nothing written or run without a card. The model is anonymous: the platform picks it.

Cost, visibly

Every turn is metered as AI credits; the footer shows N cr this session · balance · metered per turn, read on open, or RoRo Copilot is free right now in free mode. No per-message ceiling: the cost trail, Stop and your balance are the controls. Out of credits, the server refuses the next request uncharged and a Top up in the console card appears; after 40 tool rounds it pauses and Continue resumes.

Conversations and asks

Sessions are saved per project (Conversations…, +, the bin); switching rails never cuts a streaming answer, and a card waiting behind a hidden rail earns a 1 badge on the Copilot icon and a status-bar chip — click either; the rail never opens itself. Three starter chips, or a persona (hackathon, integration, research). Good asks: "What would 4000 shots cost on the cheapest real machine?", "Why did my last run fail?" — then read the diff before Apply.

Related: Python in the browser · Machines & credits · Run a circuit · Library