Skip to content

Experimental AI Chat

Open Experimental AI

Experimental

AI answers may be incorrect. Verify them against the schedule before use.

The hosted assistant is an API-key-gated beta. See AI Beta Access to request access.

This page answers questions about the schedule currently open in the browser. When enabled by the AI backend, a question can include PNG, JPEG, and WebP images or TXT, Markdown, CSV, PDF, and XLSX documents. The assistant can propose changes to the schedule, which apply only after you approve them. It cannot run optimization.

Review a proposed change

Ask for a change, such as renaming a person or adding a shift request, and the assistant proposes one instead of changing the schedule itself.

  1. Ask for the change in the usual way.
  2. Read Proposed schedule change, which lists every difference between your current schedule and the proposal.
  3. Select Approve to apply it, or Reject to discard it.
  4. An approved proposal replaces the schedule in one step, so Ctrl+Z reverts the whole change.

See what the assistant did

Small grey rows under an answer record how it was produced. They stay collapsed until you select one.

  • Reasoning shows the model's own thinking for that step.
  • The bash tool shows the command the assistant ran and its bounded output. The assistant may read task-sized schedule references from the temporary sandbox when it needs schema guidance.
  • schedule edit shows the changed lines after a Bash command changes the temporary schedule. Red lines were removed and green lines were added. This is a preview only. The current schedule still changes only after you approve the final proposal.
  • A row marked failed means that step changed nothing. This is the usual reason an answer arrives without a proposal.

Long output is revealed a portion at a time with Show more. Clear Show reasoning or Show tool activity near the top of the page to hide either kind. The choice is remembered on this browser.

Approval is refused when the schedule changed after the proposal was made. Ask again so the assistant works from what you now have.

Real scenario example

For the anonymized 87-person ward example, load large-ward-with-87-people-2025-11.yaml, then open Experimental AI. Confirm that the snapshot shows 87 people and 30 dates before asking a question.

Experimental AI chat with the 87-person schedule loaded

The blue development-version banner appears only in non-release builds.

Ask about a schedule

  1. Finish editing or load the intended schedule.
  2. Open Experimental AI.
  3. Confirm the displayed people and date counts.
  4. If Attach files is available, optionally select supported images or documents and confirm the displayed files. Remove an incorrect file with its × button.
  5. Enter a question and select Send. Press Enter to send or Shift+Enter for a new line. Where browser speech recognition is available, select the microphone to dictate the question.
  6. While the assistant is working, enter another message and select Queue to steer the same turn after its next tool call. This does not cancel work already in progress.
  7. Select Stop to interrupt a response in progress.
  8. If you scroll up in a long conversation, use the floating down-arrow button to return to the composer and latest message.

The animated Thinking indicator means the assistant is waiting for its first response text.

Assistant answers render Markdown, including headings, lists, links, code, and tables. Use the copy icon at the top-right of a code block to copy its contents. Raw HTML is ignored. Remote images written in an answer are not loaded.

Ask how to use the app

The assistant runs inside the existing Nurse Scheduling app and can explain which page and visible control to use for a task. For example, ask how to add a person, upload schedule YAML, configure a rule, or start optimization. It can guide you through those controls, but it cannot navigate, click, upload, or run optimization for you.

Files attached with Attach files belong to the next chat message. To replace the schedule currently open in the app, use Upload on Save and Load instead.

The browser uploads one YAML snapshot when it creates the chat session. Later questions in that session use the same backend-owned snapshot. Reload the page to begin a new chat from the latest schedule.

Data and limitations

The complete schedule YAML is sent to the configured AI backend and model provider. Attached images are also sent to that provider for the current question, as are the extracted contents and filenames of attached documents. PDF extraction reads embedded text and does not perform OCR. XLSX extraction includes cell values, formula text, and cached formula results when available. It does not recalculate formulas.

Assume all AI chats and related data are logged and may be used to improve this product and the AI provider's product. The AI route does not currently anonymize schedules, chats, or attachments. Do not submit personal, confidential, regulated, or otherwise sensitive information. See the privacy policy for details.

Chat sessions use unguessable identifiers but do not have account authentication yet. Model answers may be incorrect, so verify them against the schedule.