Media Studio

Start with the enabled workflow

Check which generation options your deployment exposes. A visible interface does not imply every model, output format, or external service is configured.

Match the task to the environment

Generation workloads can have different model, storage, and compute needs from text conversation. Plan around the actual workflow and supported configuration.

Understand provider boundaries

Some capabilities may rely on configured external services. Review where inputs are sent and where outputs are stored before using sensitive material.

Review the result

Check the completed artifact and its suitability for use. A submitted job, a stored file, and an accepted final result are separate milestones.

Plan, generate, inspect

Open Media Studio from the workspace layout controls. Choose only a workflow enabled in your deployment. Describe the intended output, dimensions or duration where supported, and source material. Review the actual result for content, quality, and suitability before sharing it.

Ask Friday to guide you

In an authorized FridayLocalAI conversation, try:

Help me prepare a brief for the Media Studio workflow I have available: [workflow]. Ask about purpose, audience, output, style, and source material. Explain what I should check after generation. Do not claim a job ran or a file exists until verified.

Tell Friday what you can see and which step you have reached. If the instructions do not match your screen, say so; ask for the relevant Help topic or contact your operator. Friday cannot be assumed to see your screen or know your current settings.

Start with a creative brief

Before opening a generation workflow, describe the artifact you want to make and how it will be used. Identify the audience, subject, style, and practical output requirements. A small, clear brief is easier to evaluate than an open-ended request for something “amazing.” If the workflow exposes dimensions, duration, or other settings, connect those choices to the intended use. Ask Friday to help refine the brief in a conversation. It can question unclear requirements and separate essential details from preferences. Make clear whether you want planning help or an actual generation attempt. A written prompt is not evidence that a job has been submitted, and a visible Media Studio panel is not proof that every possible media service is configured.

Check the available workflow

Open Media Studio using the workspace layout control. Inspect the options exposed by your deployment and role. The interface is intended for configured image, video, and audio workflows, but actual availability depends on services, models, providers, and operator setup. Choose the workflow that is both available and appropriate to the artifact you need. If an expected option is missing or unavailable, ask the operator rather than assuming a conversational instruction can enable it. Describe the intended result and what the interface shows. Keep this distinction especially clear when comparing a public description with an early-access installation: the description explains the product surface, while the actual configuration determines what can run in that environment.

Understand the input boundary

Reference material can help define a task, but it should be material you are permitted to use. Confirm the processing path before supplying sensitive or proprietary inputs. A local-first workspace may use a configured external provider for a particular media capability, so do not assume the same data path as a local text conversation. State how a reference should be used: preserve an approved layout, explain a subject, or guide a style choice. Do not let an ambiguous reference stand in for a complete brief. If the input cannot be read by the selected workflow, identify that limitation rather than allowing a guessed interpretation to proceed. The question is what the generation process actually receives, not merely what is stored in the workspace.

Make the first attempt reviewable

Use a bounded initial request that tests the essential idea. Decide in advance which characteristics matter most, such as the subject, composition, legibility, or relation to the brief. Avoid changing every setting at once after an unsatisfactory result, because that makes it harder to understand which change helped. When a job is submitted, inspect the available status and wait for the relevant completion evidence. A submitted request, a running job, a stored output, and an accepted final artifact are separate stages. If the process fails or stops, preserve the visible status before trying again. A retry should account for whether an output already exists and whether another attempt is appropriate for the configured service.

Review the artifact itself

Open the completed result and assess it against the brief. Do not rely only on a thumbnail, a filename, or a message saying the task is finished. Check content, dimensions or duration where relevant, quality, and suitability for the intended audience. If text appears inside an image or video, inspect the spelling and legibility directly. For revisions, describe the specific discrepancy and the desired change. “The title is unreadable at the intended display size” gives a clearer target than “make it better.” Keep approved elements and requested changes explicit. Generated media may need further review before publication or sharing, and a successful technical output does not settle whether the artifact meets the creative or practical requirements of the task.

Preserve the useful result and its context

Keep the accepted output associated with a clear description of its purpose and the decisions that produced it. Use the storage and export workflow supported by your deployment. Verify that the file you intend to use is the one you actually open or download, especially when several revisions have similar names. If the result will be shared, perform the review appropriate to that use and confirm any relevant source permissions. Friday can help draft captions, a review checklist, or a concise revision brief, but it should not claim an artifact exists or was published without evidence. Media Studio is most useful when planning, configured execution, artifact inspection, and the final human decision remain distinct and understandable parts of the creative process.