AI Character Design Sheet & Consistency Studio

Build a reusable character visual pack, then use its accepted anchor and references to create new poses, expressions, and story scenes. Use this Skill for an original character, comic or game concept, brand mascot, illustrated content series, or a character that needs a coherent visual foundation across new images.

Use this package when the user wants to choose an anchor and carry a character brief into later poses, expressions, or scenes. Route a one-off illustration, portrait, or unrelated concept image with no reusable character-asset goal to beatra-ai-image-studio instead.

Inputs and routes

Start from either:

• one to four accessible ordered reference images of the same character; or
• an original character brief with appearance, style, role, and first-use scene.

Reuse any name, audience, medium, visual style, palette, costume, props, destination, and user-named must-keeps already in the conversation. Inspect only reference images the host can actually view and record their visible role: front view, side or three-quarter view, full body, expression, outfit, prop, or style. If an image is not visible to the host, do not claim to have inspected it; record the user's stated role instead. Upload is transport only, not inspection; retain every returned artifact reference.

For an original character, create a small set of anchor concepts first and ask the user to choose one accepted anchor before making follow-on scenes. For an existing character, use ordered references to guide a new pose, expression, or scene. Use an accepted anchor as the base image only when the user wants a focused local adjustment; otherwise create a new composition guided by the ordered reference set.

Golden path

1. Form a character brief that separates identity traits from the requested scene. Capture face and silhouette, hairstyle, costume, palette, signature props, style, mood, destination, and the must-keeps that matter most. Read character profile and references for source roles and profile reuse.
2. Choose beatra.images.generate for an original anchor, beatra.images.transform for a new scene guided by one to four ordered references, or beatra.images.edit for a focused adjustment of an accepted anchor. Read the current model card for the selected capability before setting the canvas, count, relationship, model, or controls.
3. Show the selected route, reference order and roles, character brief, scene brief, must-keeps, canvas, output count, model behavior, and current maximum price. Planning a character sheet or reviewing references is free; a paid image request begins only after the user approves the frozen plan.
4. Create one opaque, stable client_request_id for the exact paid request. Submit it once with the package's bundled scripts/mcp_client.py, retain the returned task ID, and poll that same task to a terminal result.
5. Review accessible results against the brief's named must-keeps: visible identity, silhouette, costume, palette, signature elements, expression, scene direction, and destination fit. Report any observed visual drift; references guide the next image but never justify a promise of pixel-level or cross-generation absolute consistency. Only when the user asks to retain the profile and names a project location, save accepted anchors and references there. Otherwise deliver their artifact references and order for the next scene. A focused follow-up becomes a new paid request with its own approval and ID.

Paid changes, recovery, and delivery

One concept set or one scene card is one explicit paid request. When the user chooses a 1–4 image set, confirm its exact count and maximum price before submitting it. A changed character brief, reference image or order, anchor, scene, canvas, count, relationship, model, seed, palette, or edit region is new paid work with a new stable ID and approval.

Record the frozen payload, approval, create response, task ID, and terminal result. If a create response is lost, retry only the identical payload with the same ID. If the task ID is unavailable, use beatra.tasks.list and beatra.tasks.get to recover the matching task before considering another call. Queued and running tasks remain the original work. Call beatra.tasks.cancel only when the user asks, then verify the resulting task state before planning a replacement.

Deliver the returned image artifacts and only observed result facts, including dimensions, format, resolved model, successful-image count, and billing.net_charged_credits when returned. State what the host could inspect and any observed drift from the must-keeps. Do not imply that Beatra stores a character project: retain a reusable profile only in the user-selected location after explicit approval; otherwise provide the accepted artifact references and their ordered roles for a later request.

Execution

Invoke every remote Beatra operation only through this package's bundled scripts/mcp_client.py. Put the MCP tool name after call and pass its JSON arguments on standard input:

printf '%s' '{"capability":"image_to_image"}' | python3 scripts/mcp_client.py call beatra.models.list
printf '%s' '{"prompt":"Create a character anchor from the approved brief.","count":1,"client_request_id":"opaque-character-id"}' | python3 scripts/mcp_client.py call beatra.images.generate

Do not configure or call a host Beatra Connector, and do not use REST/OpenAPI as a fallback.

References by task

• For reference order, character traits, anchor selection, and reusable profiles, read character profile and references.
• For route-specific requests, approval, task polling, recovery, cancellation, and result review, read character-image workflow.
• For authorization and the non-billable registration step, read installation and authentication and installation registration.
• For shared task, billing, and connection details, read tasks and results, billing, errors, and recovery, and Bundled MCP Client diagnostics.
• For update guarantees and controls, read automatic updates and safety. For removal, read uninstall and disconnect.

Runtime and safe automatic updates

The bundled client silently checks for a newer release at most once every 24 hours per installation. When a higher version is available, it installs automatically without separate confirmation. It downloads only from the fixed official Beatra discovery and immutable CDN paths for this package, channel, and locale, verifies discovery data, archive, manifest, and every packaged file, and replaces only package-owned files.

Update checks, downloads, verification, replacement, rollback, and recovery fail open: the current installation remains usable and the original command continues. An update failure never authorizes retrying a paid image request. The setting persists for this installation. See automatic updates and safety.

python3 scripts/mcp_client.py update --auto off
python3 scripts/mcp_client.py update --auto on
python3 scripts/mcp_client.py update --check