Final print size and viewing distance must be known or assumed.
Bleed and safe margin cannot be optional.
Small details must have minimum physical size.
Sponsor / QR / legal blocks need a controlled zone.
Social crop variants need recomposition.
Image resolution must match print scale.
Dark backgrounds and subtle gradients require print proof.
Output Loss Layer
What degrades first: 小字、低对比副标题、细线、暗部图像细节。
What collapses under pressure: detail hierarchy、metadata grouping、quiet focal points。
What survives: 大标题、强形状、明确色块、简单符号。
What becomes noise: 多余纹理、装饰图形、赞助 logo 群、复杂背景。
What loses brand signal: 被活动信息和赞助信息挤压的品牌资产。
Failure Under Execution
设计稿在屏幕上成立,印成 A2 后副标题和时间地点不可读。
主视觉在 3m 外好看但无法识别主题。
社媒裁切后标题断裂,海报 stop point 消失。
暗色背景印刷后糊成一片,图像层次丢失。
赞助 logo 增加后破坏底部信息区,视觉重心下坠。
二维码和时间地点抢走主标题注意力。
Production Recovery Rules
If distance readability fails, enlarge title and simplify image before increasing contrast effects.
If sponsor clutter appears, create a locked sponsor zone and reduce its visual weight.
If print glare weakens contrast, flatten background and increase type weight.
If social crop fails, create separate crop-specific layout, not automatic resize.
If details are unreadable, group metadata and remove nonessential descriptors.
If dark print loses depth, reduce tonal complexity and protect foreground contrast.
If QR code competes, isolate it with whitespace and lower surrounding emphasis.
Injected Cases
Case Name: Swiss poster hierarchy principles
Type: canonical
Why It Matters: 建立主标题、构图和信息组的秩序。
What It Validates: poster hierarchy must be explicit。
What It Corrects: 纠正信息等权。
What It Adds: distance-aware hierarchy。
Case Name: Typography readability principles
Type: canonical
Why It Matters: 支持观看距离和标题识别。
What It Validates: experimental type needs readability boundary。
What It Corrects: 纠正不可读实验字体。
What It Adds: type-distance judgment。
Case Name: Visual semiotics in cultural posters
Type: canonical
Why It Matters: 支持符号和主题之间的关系。
What It Validates: stop point must carry meaning。
What It Corrects: 纠正无意义冲击。
What It Adds: meaningful visual hook。
Case Name: Pentagram cultural poster work
Type: studio
Why It Matters: 展示 typographic poster 的清晰与文化性。
What It Validates: restrained poster can still stop attention。
What It Corrects: 纠正噪音式冲击。
What It Adds: cultural restraint。
Case Name: Studio Dumbar dynamic poster-like systems
Type: studio
Why It Matters: 展示实验和系统控制的关系。
What It Validates: motion / dynamic logic can inform static tension。
What It Corrects: 纠正随机实验。
What It Adds: controlled energy。
Case Name: Base Design cultural typography
Type: studio
Why It Matters: 展示 quiet typography 的主题表达。
What It Validates: hierarchy can be quiet but strong。
What It Corrects: 纠正空白无焦点。
What It Adds: restrained attention。
Case Name: D&AD poster / campaign benchmarks
Type: contemporary
Why It Matters: 提供概念和传播力 benchmark。
What It Validates: idea must be quickly grasped。
What It Corrects: 纠正只看形式。
What It Adds: concept scrutiny。
Case Name: Social event poster platform patterns
Type: contemporary
Why It Matters: 提醒 poster 经常被二次裁切为移动端图。
What It Validates: crop and thumbnail testing。
What It Corrects: 纠正只为印刷或只为手机设计。
What It Adds: dual-context check。
Extracted Heuristics
Learn: poster starts with viewing distance;Strengthens: context diagnosis;Apply: every poster brief。
Learn: one stop point beats multiple medium hooks;Strengthens: focal control;Apply: event and cultural posters。
Learn: experimental type needs a readability boundary;Strengthens: type judgment;Apply: cultural posters。
Learn: details must be grouped, not scattered;Strengthens: information reduction;Apply: event posters。
Learn: quiet posters still need tension;Strengthens: taste judgment;Apply: minimal poster directions。
Learn: social crop is not the same as poster composition;Strengthens: production adaptation;Apply: digital poster variants。
Learn: symbol must carry theme, not just intrigue;Strengthens: semiotic fit;Apply: concept posters。
Learn: print details require distance and substrate checks;Strengthens: production readiness;Apply: physical posters。
Case-backed Refinements
Reinforced: stop point must be defined before layout details.
Corrected: visual impact cannot be measured only by saturation or scale.
Narrowed: experimental poster language is acceptable only when theme recognition survives.
Evaluation System: 1-5 scoring with poster-specific dimensions。
Version
Version: V1.0 (production-proven)
Status: production-proven
Last updated: 2026-05-02
1---2name: poster-design3description: Poster Design Skill4---5# Poster Design Skill67## Skill Name89poster-design1011## Purpose1213用于海报设计的传播判断:停留效率、主视觉抢占、主题优先、信息减法和传播识别。1415## When to Use1617- 活动海报、文化海报、展览海报、产品海报、发布会海报。18- 需要在短时间内吸引注意并传达主题。19- 需要判断海报主视觉、标题和信息密度是否成立。20- 需要将 poster 适配 print、digital、OOH 或 social crop。2122## When Not to Use2324- 任务是长期品牌系统时,使用 `brand-design`。25- 任务是 campaign 主视觉系统时,使用 `campaign-kv`。26- 任务只是社媒封面时,使用 `social-media-visual`。27- 任务只是字体排版细节时,使用 `typography-layout`。2829## Inputs3031- Event / product / topic32- Core message33- Required text34- Audience and viewing context35- Format, size, distance36- Brand assets37- Image or graphic assets38- Print / digital constraints3940## Outputs4142- Poster objective43- Attention hook44- Hierarchy plan45- Main visual direction46- Information reduction plan47- Type / image / color logic48- Production specs49- Evaluation score5051## Core Judgment5253先用真实观看场景判断距离、停留时间和媒介,再用 poster、typography、cultural design 案例经验判断 stop point,最后压缩信息密度并保护主题识别。5455## Decision Logic5657- First decision: 确定观看场景:远距离、近距离、手机缩略图、展墙、户外或社媒。58- Second decision: 定义一个 stop point:标题、图像、符号、冲突、色块或构图张力。59- Third decision: 划分信息密度:hero message、event details、support info、optional info。60- Evidence calibration: poster 需要同时通过 attention、theme recognition 和 readable hierarchy 的验证。61- Experience rule: 文化海报案例显示,视觉实验只有在主题识别仍然成立时才有效。62- Experience rule: 活动海报案例显示,客户给出的全部信息不能等权进入海报,必须按观看距离裁剪。63- Prioritize: stopping power、theme recognition、readability、memorability、production fit。64- Reduce: 长句、解释性副标题、赞助信息视觉权重、无关图形、过多日期格式。65- Suppress: 抢主标题的 secondary info、复杂背景、无法远读的小字、装饰性 AI 图。66- What breaks the output: 主题不清;远看没有停留点;近看信息混乱;主视觉与标题说不同事情。6768## Workflow69701. Diagnose: 判断海报类型、观看距离、停留时间和媒介。712. Define Hook: 写出海报的一个主停留点。723. Prioritize Info: 把内容分成 title、hook、details、metadata、remove。734. Structure Layout: 决定主视觉位置、标题比例、信息组和留白。745. Validate: 对照 typography、grid、semiotics 和 contemporary poster sources 校准。756. Direct Visual: 定义图像 / 图形 / 字体 / 色彩如何服务主题。767. Reduce: 删除或压低不影响传播的内容。778. Test: 用 thumbnail、3-second read、print distance 检查。789. Output: 给出可执行海报方向和评分。7980## Output Format8182- Receiver: 交付给 designer / printer / event producer / media operator 的执行对象。83- Poster Type and Context: 使用场景、距离、媒介。84- Core Message: 海报必须传达的一句话。85- Stop Point: 吸引停留的视觉机制。86- Hierarchy Plan: title / visual / details / metadata 的权重。87- Main Visual Direction: 图像或图形如何抢占注意。88- Typography Direction: 字级、字重、可读距离。89- Information Reduction: 删除、合并、弱化清单。90- Validation Binding: canonical / studio / contemporary references。91- Production Spec: 尺寸、比例、出血、色彩模式、裁切。92- Adaptation Notes: print、OOH、social crop、event signage 的差异。93- Evaluation Score: 按本 Skill 评分。9495## Quality Bar9697海报必须有一个强停留点,一个清楚主题,一条稳定阅读路径,并在目标尺寸和距离中可读。9899## Anti-Patterns100101- No stop point: 远看没有任何元素能让人停下;poster validation 必须先看停留效率。102- Multiple posters in one: 一个版面塞入多个概念,主题被稀释。103- AI wallpaper poster: 用 AI 图做背景,文字后贴上去,图文没有关系。104- Detail dominance: 时间、地点、嘉宾、赞助等细节压过主题。105- Illegible expression: 实验字体或扭曲排版让主题不可读。106- Fake impact: 只靠高饱和色、爆炸构图或粗暴对比制造刺激。107- Social-only poster: 只适合手机截图,不适合实际海报尺寸或印刷。108- Case-backed evidence: Pentagram cultural posters 显示 typographic restraint 可以比图像噪音更有传播力。109- Case-backed evidence: Studio Dumbar dynamic systems 显示实验必须由系统控制,否则变成视觉噪音。110- Case-backed evidence: Base Design cultural typography 显示安静海报也需要 strong hierarchy。111- Case-backed evidence: D&AD poster / campaign work 显示强概念必须能被快速识别。112- Case-backed evidence: social poster patterns 说明手机缩略图有效不代表实体海报有效。113114## Evaluation Criteria1151161-5 分:117118- Stopping efficiency: 3 秒内是否能吸引停留。119- Theme recognition: 主题是否被快速识别。120- Hierarchy control: 主视觉、标题、细节权重是否正确。121- Information density: 信息是否足够但不过载。122- Visual memorability: 是否有可记住的视觉机制。123- Type and distance readability: 字体是否适合观看距离。124- Brand / context fit: 是否适合品牌、活动和受众。125- Real-world performance: 远距离、近距离、手机缩略图三个场景中的主题损耗。126- Real-world performance: 印刷尺寸中细节、色彩和小字是否保持可用。127- Distance retention: 3m / 5m 观看距离下 title 和 stop point 是否保留。128- Print fidelity: 纸张、反光、CMYK 偏色后主题是否保留。129- Crop survival: 社媒裁切后主视觉和标题是否仍成立。130- Production readiness: 是否能印刷、裁切和数字发布。131132## Production Notes133134必须作为 handoff 记录:尺寸、比例、观看距离、bleed、safe margin、最小字号、CMYK/RGB、纸张、覆膜 / 反光风险、展墙 / OOH 环境、社媒裁切版本、导出格式、版权素材风险和打样检查点。135136## Canonical References137138- Typography: 支持标题、细节和观看距离可读性。139- Grid & Layout: 支持主视觉、信息组和留白比例。140- Visual Semiotics: 支持符号、主题和文化识别。141142## Studio References143144- Pentagram: 可验证 cultural / typographic poster discipline。145- Studio Dumbar: 可验证 dynamic and experimental poster-like systems with control。146- Base Design: 可验证 restrained cultural typography and layout。147148## Validation Notes149150- 案例验证结论:Pentagram / cultural poster cases 强化了“停留点先于细节”的规则。151- 案例验证结论:Studio Dumbar / dynamic identity cases 修正了“实验越强越好”的误判,实验必须有系统边界。152- 案例验证结论:Base Design / typographic cases 证明 restraint 也需要 clear hierarchy。153- Contemporary Validation Sources: D&AD、Creative Review、It’s Nice That 可作为 contemporary poster / cultural design validation。154155## Evidence Gaps156157- 缺生产测试证据:3m / 5m / 手机缩略图三种观看条件下的信息保持率。158- 缺印刷 proof:覆膜反光、CMYK 偏色、纸张吸墨对标题和图像的影响。159- 缺活动现场 proof:赞助、二维码、时间地点增加后如何保持主题优先。160161## Production Proof162163First production proof focus: distance readability and print / crop survival.164165本 Skill 的第一轮生产验证关注:海报在 3m / 5m 观看距离、印刷反光、展墙环境和社媒裁切后,stop point、主题、标题和关键信息是否仍然成立。166167## Real-world Stress Conditions168169- 3m / 5m distance: 标题和主视觉必须先被读到。170- Print glare: 覆膜或灯光反射降低可读性。171- Low-light venue: 展厅或活动现场光线不足。172- Sponsor clutter: 赞助 logo 和 metadata 临时增加。173- Social crop: 海报被裁成 1:1、4:5、9:16。174- CMYK shift: 高饱和色和暗部印刷偏移。175- Wall competition: 展墙或街面周围海报竞争注意力。176177## Production Constraints178179- Final print size and viewing distance must be known or assumed.180- Bleed and safe margin cannot be optional.181- Small details must have minimum physical size.182- Sponsor / QR / legal blocks need a controlled zone.183- Social crop variants need recomposition.184- Image resolution must match print scale.185- Dark backgrounds and subtle gradients require print proof.186187## Output Loss Layer188189- What degrades first: 小字、低对比副标题、细线、暗部图像细节。190- What collapses under pressure: detail hierarchy、metadata grouping、quiet focal points。191- What survives: 大标题、强形状、明确色块、简单符号。192- What becomes noise: 多余纹理、装饰图形、赞助 logo 群、复杂背景。193- What loses brand signal: 被活动信息和赞助信息挤压的品牌资产。194195## Failure Under Execution196197- 设计稿在屏幕上成立,印成 A2 后副标题和时间地点不可读。198- 主视觉在 3m 外好看但无法识别主题。199- 社媒裁切后标题断裂,海报 stop point 消失。200- 暗色背景印刷后糊成一片,图像层次丢失。201- 赞助 logo 增加后破坏底部信息区,视觉重心下坠。202- 二维码和时间地点抢走主标题注意力。203204## Production Recovery Rules205206- If distance readability fails, enlarge title and simplify image before increasing contrast effects.207- If sponsor clutter appears, create a locked sponsor zone and reduce its visual weight.208- If print glare weakens contrast, flatten background and increase type weight.209- If social crop fails, create separate crop-specific layout, not automatic resize.210- If details are unreadable, group metadata and remove nonessential descriptors.211- If dark print loses depth, reduce tonal complexity and protect foreground contrast.212- If QR code competes, isolate it with whitespace and lower surrounding emphasis.213214## Injected Cases215216- Case Name: Swiss poster hierarchy principles217 Type: canonical218 Why It Matters: 建立主标题、构图和信息组的秩序。219 What It Validates: poster hierarchy must be explicit。220 What It Corrects: 纠正信息等权。221 What It Adds: distance-aware hierarchy。222- Case Name: Typography readability principles223 Type: canonical224 Why It Matters: 支持观看距离和标题识别。225 What It Validates: experimental type needs readability boundary。226 What It Corrects: 纠正不可读实验字体。227 What It Adds: type-distance judgment。228- Case Name: Visual semiotics in cultural posters229 Type: canonical230 Why It Matters: 支持符号和主题之间的关系。231 What It Validates: stop point must carry meaning。232 What It Corrects: 纠正无意义冲击。233 What It Adds: meaningful visual hook。234- Case Name: Pentagram cultural poster work235 Type: studio236 Why It Matters: 展示 typographic poster 的清晰与文化性。237 What It Validates: restrained poster can still stop attention。238 What It Corrects: 纠正噪音式冲击。239 What It Adds: cultural restraint。240- Case Name: Studio Dumbar dynamic poster-like systems241 Type: studio242 Why It Matters: 展示实验和系统控制的关系。243 What It Validates: motion / dynamic logic can inform static tension。244 What It Corrects: 纠正随机实验。245 What It Adds: controlled energy。246- Case Name: Base Design cultural typography247 Type: studio248 Why It Matters: 展示 quiet typography 的主题表达。249 What It Validates: hierarchy can be quiet but strong。250 What It Corrects: 纠正空白无焦点。251 What It Adds: restrained attention。252- Case Name: D&AD poster / campaign benchmarks253 Type: contemporary254 Why It Matters: 提供概念和传播力 benchmark。255 What It Validates: idea must be quickly grasped。256 What It Corrects: 纠正只看形式。257 What It Adds: concept scrutiny。258- Case Name: Social event poster platform patterns259 Type: contemporary260 Why It Matters: 提醒 poster 经常被二次裁切为移动端图。261 What It Validates: crop and thumbnail testing。262 What It Corrects: 纠正只为印刷或只为手机设计。263 What It Adds: dual-context check。264265## Extracted Heuristics266267- Learn: poster starts with viewing distance;Strengthens: context diagnosis;Apply: every poster brief。268- Learn: one stop point beats multiple medium hooks;Strengthens: focal control;Apply: event and cultural posters。269- Learn: experimental type needs a readability boundary;Strengthens: type judgment;Apply: cultural posters。270- Learn: details must be grouped, not scattered;Strengthens: information reduction;Apply: event posters。271- Learn: quiet posters still need tension;Strengthens: taste judgment;Apply: minimal poster directions。272- Learn: social crop is not the same as poster composition;Strengthens: production adaptation;Apply: digital poster variants。273- Learn: symbol must carry theme, not just intrigue;Strengthens: semiotic fit;Apply: concept posters。274- Learn: print details require distance and substrate checks;Strengthens: production readiness;Apply: physical posters。275276## Case-backed Refinements277278- Reinforced: stop point must be defined before layout details.279- Corrected: visual impact cannot be measured only by saturation or scale.280- Narrowed: experimental poster language is acceptable only when theme recognition survives.281282## Inherits283284- Canonical System: hierarchy、contrast、composition、type readability。285- Trend Overlay: 只判断表达是否过时或饱和。286- Taste System: bold vs noisy、minimal vs empty、tension。287- Anti-Pattern System: overdesign、AI wallpaper、template poster。288- Critique System: priority-based poster findings。289- Evaluation System: 1-5 scoring with poster-specific dimensions。290291## Version292293- Version: V1.0 (production-proven)294- Status: production-proven295- Last updated: 2026-05-02
Run npx skillmds@latest add rickidea0320-cpu/poster-design in your terminal (requires Node.js), paste this page's agent-chat prompt into Claude, Cursor, or any MCP-connected agent, or download the SKILL.md file and copy it into your agent's skills directory.
Poster Design Skill It is listed under Design & Media on SkillMD.
This skill has not completed SkillMD's automated safety review yet. Capability flags: docs only. SkillMD never runs a skill's scripts for you; review the SKILL.md before installing.
This skill is tagged as working with Claude Code, Claude.ai, OpenAI Codex. SKILL.md is an open format, so most agents that read a skills directory can load it too.
Yes. Installing skills from SkillMD is free, and the skill stays under its author's original license.
rickidea0320-cpu (@rickidea0320-cpu) published this skill. Their other Agent Skills are listed on their SkillMD profile.