Create and refine visuals for ads, social posts, product imagery, logos, charts, and related design work. Compare distinct directions on a shared board in the Creative Production app, then select and revise preferred results.
中文简介正在准备中。当前显示英语简介。
Use when the user wants to create, explore, adapt, refine, polish, or review visual creative such as campaigns, ads, social posts, product imagery, scenes, offers, logos, brand systems, styles, charts, decks, or related marketing and design assets.
安装前可以查看交给 AI 代理的指令内容。
Produce is the main Creative Production workflow. It turns a concrete brief, source asset, selected board item, or starter follow-up into reviewable visual work inside the Creative Production app.
intake to mount the zero-state board. Do not use structured intake.creative_production_board directly once with action=open. Do not invoke it through functions.exec; the direct MCP result carries the UI resource the host needs. Keep the returned boardId for the entire workflow.action=begin_generation and stable placeholder IDs. After each generated file exists, call action=complete_generation with the matching item ID and absolute imagePath. Use fail_generation for failed outputs. Never call open again to refresh or continue a board.references/board-runtime.md when creating, importing into, or refreshing a board.NEEDS_COORDINATOR_INPUT; close terminal workers before replacements or retries. If runtime capacity is lower than expected, reduce the wave size. Validate that every returned path exists and is non-empty. Preserve successful and cached outputs. Retry only outputs that are missing or invalid, with at most 2 total attempts per output. Do not use the local /api/images endpoint for native generation. Do not use codex exec, restore runtime/codex_exec_image_batch.py, or introduce another process-based native fallback.boardId, item ID, and revision. Every result identifies the same widget resource and board session so the host can update the existing mounted board; the bounded refresh watcher is recovery, not the primary delivery path. Do not render or remount another board.Load one or more mode references only when the request needs their distinct prompt or artifact guidance:
references/modes/ads.md for paid social, display, OOH, ecommerce, UGC, proof-led, or campaign ad directions.references/modes/scenes.md for product, service, venue, or offer placement in realistic contexts.references/modes/offers.md for offer-led, product-led, service-led, hero, or landing-page visual families.references/modes/shots.md for angles, crops, pans, zooms, macro details, or camera variants from an anchor image.references/modes/logos.md for new identity directions, marks, wordmarks, and lockups.references/modes/styles.md for reusable palette, typography, texture, lighting, and visual-system directions.references/modes/positioning.md for audience, occasion, proof, differentiation, and visual implication exploration.references/modes/charts.md for chart-specific visual treatment that preserves data and meaning.Load cross-cutting contracts only when their condition applies:
references/review-renderer.md when a helper produces a review manifest or local review page.references/contracts/source-preservation.md when an existing image, selected board item, product, person, packaging, logo, or chart must remain recognizable.references/contracts/exact-content.md when copy, chart values, labels, logos, claims, dimensions, or legal content must remain exact.references/contracts/deterministic-exports.md when the user asks for publish-ready, platform-specific, or final exported assets.When no mode requires a different structure, generate 4-6 distinct visual directions. Each board tile must be one clean visual asset, not a collage, contact sheet, or mood board inside a mood board.
Use safe defaults for sparse but concrete requests. Do not invent claims, prices, certifications, endorsements, product facts, or readable copy.
Lead with the active board and what the user should compare, select, reject, or refine next. Do not lead with local files, manifests, server URLs, or implementation details.
还没有评价。欢迎分享你的使用感受。
无需登录即可发表。欢迎分享使用感受、提问，或说明在哪些环境下无法运行。
还没有评论。来写第一条吧。
Review charts, diagrams, and dashboards for accessibility, including color contrast, keyboard access, and text alternatives. Get specific guidance for preserving key findings in screen readers, static exports, and reduced-motion views.
中文简介正在准备中。当前显示英语简介。
Give a photo collection a consistent look using Adobe tools to adjust color, tone, presets, and optional crops. Review a before-and-after sample before processing the full set, then receive links to the edited images.
中文简介正在准备中。当前显示英语简介。
Visualize a logo, sketch, or design on products such as mugs, shirts, and business cards. Review a pilot image, generate a visually consistent set, and collect the results on a shareable Firefly Board.
中文简介正在准备中。当前显示英语简介。
Prepare images and videos for social platforms using Adobe tools. Create image crops that keep the subject in view, expand backgrounds when available, preview a full-set sample, and resize videos while preserving their aspect ratio.
中文简介正在准备中。当前显示英语简介。
Create flyers, social posts, business cards, and other visuals from Adobe Express templates. Choose a template, customize its text and background color, and receive a design link, with additional editing and export options depending on the available tools.
中文简介正在准备中。当前显示英语简介。
Turn an uploaded video into three highlight edits with Adobe Quick Cut. Choose a target length and style, then compare versions that feature different moments and download the ones you prefer.
中文简介正在准备中。当前显示英语简介。