Skip to main content

Work with tasks

Tasks group an outcome with the context and steps needed to finish it. Keep titles outcome-oriented and put supporting notes, images, recordings, and time records inside the task.

All screenshots
ScreenshotMobile view
Production task workspace showing real task rows with cover-art thumbnails from the Intergalactic Penguin Logistics demo data

On desktop, the task list remains visible beside the selected task so you can move through work without losing context. On mobile, selecting a task opens its detail view full-screen. Cover art follows the task everywhere it appears, including the task list and linked Daily OS agenda items.

Cover art is optional, but it can make a dense workspace much easier to scan. Use it for a useful visual cue, a product reference, or a memorable metaphor for the work. The Project Waddle illustrations in this manual are only an example: a cover can be as practical, playful, or wonderfully weird as you want.

All screenshots
ScreenshotMobile view
Production task detail page for inspecting the orbital penguin habitat, with the same cover art used by task thumbnails

The same task can hold a description, category, labels, due date, time records, checklist items, images, links, and audio recordings.

Assign a task agent

A task can have one task agent. Assigning one gives that task a persistent AI assistant with its own inference setup, report, wake state, and proposal history. The agent reads the task and its linked context, distils the current situation into a short report, and can surface proposed task or checklist changes for you to confirm or reject.

The assigned agent appears below the task's checklists and linked tasks. Its collapsed card keeps the latest summary, automatic-update control, Wake agent action, and active model route visible without making the task page feel like an AI console.

All screenshots
Habitat Watcher is assigned to this task; assignment and automatic updates are separate choices.
ScreenshotMobile view
Assigned Habitat Watcher task agent on the orbital penguin habitat task, showing its current summary, automatic updates, wake action, and Project Waddle model route

On a phone, the card follows a single reading order: identity, update controls, AI setup, report, then proposed changes and history. On desktop, the report and proposals form the reading column while the compact controls remain nearby.

Choose Read more to expand the complete report in place. The expanded card keeps the task visible, adds the report's supporting assessment and recommended next step, and offers Open agent internals when you need reports, conversations, observations, or runtime details.

All screenshots
The full report expands inside the task instead of sending you to a separate page.
ScreenshotMobile view
Expanded Habitat Watcher task-agent card with the full Project Waddle habitat assessment and recommended next step

Choose the agent's AI setup

The card's Current setup row identifies the profile, model, and provider used for its next wake. Select it to choose a saved inference profile, override only the thinking model, return to the profile default, copy the category's current setup, or turn off AI for this one agent. A choice stays with that agent for future updates; copying a category setup is a snapshot, not a link that silently changes later.

If a report was produced through a different route, Lotti keeps a separate This report attribution instead of relabelling history with the current setup. Turning AI off leaves the existing report readable but prevents further wakes until you select a working setup again. Automatic updates cannot be enabled while the agent has no working AI setup.

Review proposed changes

A completed wake can update the report and add proposed changes. Automatic updates decide when the agent wakes; they do not give the agent permission to edit the task. Proposals remain pending until you act on them.

All screenshots
The agent proposes a feeder-test checklist item and a revised estimate; neither change has been applied yet.
ScreenshotMobile view
Habitat Watcher task-agent card with two pending Project Waddle proposals, individual dismiss and confirm actions, and a Confirm all action

Use the cross to Dismiss a proposal or the check to Confirm and apply it. You can also swipe a proposal left to dismiss or right to confirm. When several proposals are pending, Confirm all applies the currently listed changes together. Review bulk proposals carefully, especially status, date, and checklist changes.

Open History to revisit recent outcomes. It distinguishes confirmed changes from dismissed ones. A dismissed entry can represent either your decision or a proposal the agent retracted after newer task context made it stale.

A manual Wake agent can produce the same kinds of report and proposals as an automatic wake. Turning automatic updates off stops future background wakes; it does not apply, discard, or otherwise resolve proposals that are already waiting for your decision.

Choose automatic or manual updates

New task agents start with Automatic updates off. Assignment alone does not spend tokens in the background. While updates are off, Lotti continues to notice relevant changes to the task and its linked entries, marks an older summary as out of date, and waits for you to choose Wake agent. A manual wake remains available whenever the agent has a working AI setup.

All screenshots
Manual mode observes changes and reports staleness without starting inference.
ScreenshotMobile view
Habitat Watcher task-agent card with automatic updates off, an out-of-date report notice, and a manual Wake agent action

The out-of-date notice deliberately leaves the last useful report in place. If a relevant edit arrives after a wake begins, that newer edit keeps the report marked stale even when the earlier wake finishes. When you need a report that includes those latest changes, wake the agent again after the task has settled.

When Automatic updates are on, Lotti bundles matching task changes for about two minutes before scheduling a refresh. This coalescing avoids an agent wake for every small edit. Wake agent is still available for an immediate, user-requested refresh.

Turning automatic updates off clears its pending countdown and queued automatic refreshes, but does not cancel a wake you explicitly requested. Turning it on again applies to future changes; it does not replay changes that arrived while automation was off.

Priority

Lotti uses four compact priority levels:

PriorityMeaningUse it for
P0UrgentWork that genuinely needs attention next
P1HighImportant work that should happen soon
P2MediumThe normal default
P3LowWork that can wait

Open a task and use the priority control in its header to change the level. Existing tasks without an explicit priority behave as P2.

The picker spells out the compact codes, while the priority glyphs add a visual cue in both light and dark themes. Treat the P0–P3 codes as the meaning; color helps scanning but should never be the only signal you rely on.

Filter and order the task list

Narrated walkthrough: filtering the task list by status and category, then saving the view.

Use the filter control beside the task search field to focus the list. Choose one or more statuses, priorities, categories, labels, and—when Projects are enabled—projects. Category and label selectors also offer Unassigned for work that still needs that organizing detail. The Agent choice distinguishes tasks with an assigned agent from tasks without one.

All screenshots
ScreenshotMobile view
Task filter sheet with sort, status, priority, category, and label controls

Choices within one filter are alternatives: selecting P0 and P1 shows either priority. Different filters narrow the result together: selecting P0 or P1 and Penguin Operations shows only matching work. Use Clear all to return to an unconstrained view rather than trying to undo each choice individually.

Choose Due Date, Created, or Priority in the same sheet to change the ordering. Priority order is P0 through P3, with newer tasks first within a priority unless another sort is active. On desktop, an enabled vector-search setup can also offer semantic search alongside normal full-text search.

When you regularly return to the same focused view, save the active filter from the sheet and give it a clear name. Reopen it from the task list, then update the saved filter when its criteria change.

Use checklists for the next layer down

A checklist item should be smaller and more concrete than its task. “Publish the manual” is a task; “verify every navigation link” is a checklist item. Checklists can be edited directly and may also be proposed by configured AI skills from task context.

Add task context

Use the task's Add action to attach a checklist, another task, an audio recording, a timer, a text entry, or an image without leaving the task. The available actions depend on the current platform and context.

All screenshots
ScreenshotMobile view
Add menu over the orbital habitat task with task-aware capture actions

See Add entries and context for when to use each entry type.

Generate cover art from real task imagery

Narrated walkthrough: generating cover art from task context and reference images.

If an image-generation skill is configured, open the task's generation menu and choose its cover-art action. Lotti first shows images already attached to the task, plus available cover images from linked tasks. Select up to five references to guide the result, or continue without a reference.

The generated cover is created in Lotti from the task's context and any references you select. It is a visual aid, not a required field: use it when a distinct image makes the work easier or more enjoyable to recognize. The current image-generation route uses the configured provider, so treat the generation prompt and selected references as data you are intentionally sending to that provider; a local thinking or transcription route does not make cover generation local.

All screenshots
The picker uses the same task imagery visible throughout the Project Waddle demo.
ScreenshotMobile view
Generate Cover Art sheet showing five distinct Project Waddle reference images from the task context

After you continue, image generation runs in the background. The progress view uses Lotti's production thinking visualizer and can be closed without cancelling the job.

All screenshots
ScreenshotMobile view
Task cover-art generation in progress with Lotti's production shader visualizer

The generated image becomes the task cover and follows the task into list, detail, project, and Daily OS views. A strong cover should remain recognizable when the wide source image is cropped into a compact task thumbnail. It can be literal evidence, a restrained metaphor, or as strange as the work deserves; the only useful test is whether you can recognize the task faster later.

Open the task's Linked section and choose to link an existing task. Search the picker, then select the related outcome. Lotti excludes the current task and tasks that are already linked in that direction.

All screenshots
ScreenshotMobile view
Production task-link picker listing related Project Waddle tasks and their status, priority, category, and due-date context

Task links are directional: the current task points to the task you choose. Both sides remain reachable through their linked-entry sections, while each task keeps its own status, schedule, and attachments.

Use voice without losing control

Narrated walkthrough: dictating a task, reviewing the transcript, and checking off a suggested checklist item.

Record a note inside the task when it is faster to explain the situation aloud. With transcription and task skills configured, Lotti can turn that context into text and proposed checklist changes. Review the result, especially names, dates, and destructive changes. See Record audio for the recording controls and processing options.

tip

Use P0 sparingly. If everything is urgent, the label stops carrying information.