WeChat Article Visual Pack

Create a coordinated two-to-four-image WeChat Official Account visual pack from a finished article, outline, brand assets, photos, or visual references. The pack has a lead visual followed by section-focused article illustrations with one consistent direction and reading order; it is not a replacement for one standalone article cover.

Choose the route

Lock the scope first: cover only, body illustrations only, or the full pack. A cover-only request that needs a 2.35:1 headline cover can stay on this pack.

• Create from article content: use beatra.images.generate when the title, article outline, section roles, and visual direction are sufficient.
• Compose from brand assets: upload up to four ordered photos, products, portraits, logos, or reference images and use beatra.images.transform, clearly declaring what each reference preserves or guides.
• Refine a selected image: use beatra.images.edit with the accepted lead visual or article illustration as images[0], with no more than two normalized local edit regions.

This Skill owns an ordered image set for a long-form article. Use wechat-cover-maker when the user needs one focused article cover only; it uses its own cover-specific canvas and should not be substituted for this pack.

Plan the visual sequence

The hard input is the article or outline. Do not invent illustration themes when the article is missing.

Reuse the article's message, section outline, audience, brand assets, visual references, and must-keep details. Propose two to four roles: a lead visual first, then the most important sections, process steps, examples, or evidence. Default to a coherent 2K 16:9 image system for the pack—not the existing single-cover 900 × 383 path—and include the exact count, order, canvas, and text treatment in the paid-call confirmation.

Plan one visual language: shared palette, subject treatment, lighting, and headline-safe areas. Default to one accent color and one visual system for the whole article; each image carries one concept. Do not promise exact in-image Chinese copy, logos, or line breaks. If the user requests short words in the image, freeze the exact copy in the confirmation and inspect it only when the output is visible.

Golden path

Propose the illustration outline first and wait for the user to confirm the style. Then generate paid images by position. Prefer existing business images over new generation.

Verify sequence capability, then confirm one paid request

Use only this Skill's bundled scripts/mcp_client.py for every remote Beatra operation. Do not configure or call a host Beatra Connector and do not use REST/OpenAPI as a fallback. Upload local images through the bundled client; an upload makes bytes available but does not inspect their content.

Before proposing a multi-image paid route, call beatra.models.list with the exact capability for the next tool and verify that a current route supports that capability, final count, canvas, and output_relationship: "sequence":

python3 scripts/mcp_client.py call beatra.models.list

{"capability": "text_to_image"}

For transform use {"capability":"image_to_image"}; for edit use {"capability":"image_edit"}. Call once for the chosen route.

Keep model: "auto" by default; only a real availability, compatibility, control, or price decision should select a concrete model. If sequence is unavailable, report the compatibility outcome and obtain the user's explicit choice to produce one image or split the work into several paid requests. Never silently change a requested linked sequence into independent candidate images.

When available, show a single paid-call confirmation that freezes the prompt, lead and section order, count, sequence relationship, canvas, ordered references, text treatment, model, and controls. Planning and compatibility checks are free. After approval, create one opaque stable client_request_id without user content and submit once. Any changed prompt, roles, count, canvas, references or order, relationship, model, or control is new paid work and needs a new confirmation and ID.

Track, review, and deliver

Keep the returned task_id and poll only it with beatra.tasks.get. If the ID is lost, inspect beatra.tasks.list, then verify the chosen candidate with tasks.get. Replay the exact same payload with the same ID only when creation response delivery is genuinely unknown. Slow polling, connection, update, and authorization failures never cause a replacement paid request.

Use cancellation only when the user asks. A 409 response from beatra.tasks.cancel means the original task continues. When results are visible, review returned order, lead and section focus, visual consistency, safe areas, confirmed canvas, and must-keep assets. Deliver returned artifact links in sequence, observed dimensions, task ID, resolved model, and billing.net_charged_credits; offer at most one focused unexecuted refinement.

References by task

• Sequence payloads, asset ordering, and local slide edits: workflow
• Install or reauthorize: installation and authentication
• Bundled-client diagnostics: Bundled MCP Client diagnostics
• Registration: installation registration
• Polling and result fields: tasks and results
• Billing and recovery: billing, errors, and recovery
• Disconnecting: uninstall and disconnect
• Safe updates: automatic updates and safety

Runtime and safe automatic updates

The bundled client silently checks for a newer release at most once every 24 hours while a public command runs. When a newer version is available, the client installs it automatically without separate confirmation only from fixed official Beatra discovery and immutable CDN paths. It verifies the archive, manifest, and every package-owned file, then replaces only files owned by this package. If an update step fails, the current installation remains usable and the original command continues. The setting persists for this installation.

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