agency-brand-scoping
Compare original and translation side by side
🇺🇸
Original
English🇨🇳
Translation
ChineseAgency brand scoping
代理品牌范围界定
A fast discovery pass for a new client: gather brand signals (site, deck, competitors, audience, tone), lock them into a reusable file, and produce 5 on-direction visual explorations to validate with the client before any production spend.
brand-system.jsonURL / deck in → brand-system.json + 5 direction variations out. One hour, under $1 in generation cost, reusable across the rest of the engagement.
针对新客户的快速探索流程：收集品牌信号（官网、演示文稿、竞品、受众、调性），并将其存入可复用的文件，产出5组符合品牌方向的视觉探索方案，以便在投入生产资源前与客户确认。
brand-system.json输入：网址/演示文稿 → 输出：brand-system.json + 5组方向变体。耗时1小时，生成成本低于1美元，可在整个合作项目中复用。
When to Use
适用场景
- Responding to an RFP and need 5 distinct visual directions for the first review
- Net-new client pitch — you've read the brief, now you need on-brand sketches
- Kickoff phase of a signed engagement, before production generations start
- Scoping a rebrand: grab the existing brand, propose 5 evolution paths
- Any moment where "what does this brand look like in AI generation" is the open question
Do not use for finished campaign assets — this is discovery only. Lock the system here, then run or for deliverables.
agency-pitch-mockupsagency-multi-brand-pack- 响应RFP（提案请求），需为首次评审提供5种独特视觉方向
- 新客户提案——已研读brief，需产出符合品牌调性的草图
- 已签约项目的启动阶段，正式生成生产资源之前
- 品牌重塑规划：获取现有品牌信息，提出5种演进路径
- 任何需要解答“AI生成的品牌视觉是什么样”的场景
不适用于 完成的营销活动资产——此流程仅用于探索阶段。确定品牌体系后，可运行或来生成交付物。
agency-pitch-mockupsagency-multi-brand-packPrerequisites
前置要求
Ask the user (batch in one message):
- Client name + slug — used for folder + manifest tags (e.g. )
acme-fintech - Brand references — URL, existing deck, Figma file, or "no brand yet, we're defining it"
- Competitors / comparable brands — 2-3 names; informs what NOT to look like
- Audience + tone — who buys, what feeling ("premium + restrained" vs "bold + irreverent")
- Deliverable type the scope is for — pitch deck, campaign, launch film, product shoot (informs aspect ratios)
- Confidentiality — is this NDA? If yes, never name the client in public Drive folders or prompts
If the user gives a URL or deck, read/fetch it first and extract palette, typography impression, imagery style, and tone words. Propose the back for confirmation before generating.
brand-system.json向用户一次性询问以下信息：
- 客户名称+简称 —— 用于文件夹和清单标签（例如：）
acme-fintech - 品牌参考资料 —— 网址、现有演示文稿、Figma文件，或“尚无品牌，正在定义中”
- 竞品/同类品牌 —— 2-3个品牌名称；用于明确需要规避的风格
- 受众+调性 —— 目标客户群体、想要传递的感受（如“高端+克制” vs “大胆+不羁”）
- 本次范围对应的交付物类型 —— 提案演示文稿、营销活动、启动影片、产品拍摄（影响宽高比）
- 保密要求 —— 是否签署NDA？若是，绝不能在公共云端文件夹或提示词中提及客户名称
如果用户提供了网址或演示文稿，需先读取/获取内容，提取配色方案、字体印象、图像风格和调性关键词。在生成方案前，需提交供用户确认。
brand-system.jsonHow to Run
操作流程
1. INGEST → read URL / deck / Figma, extract signals
2. DRAFT → propose brand-system.json (palette, type-feel, imagery, tone, do-nots)
3. CONFIRM → user locks the system; save to clients/<slug>/brand-system.json
4. ESTIMATE → gen-ai pricing on the 5-direction batch (< $1 target)
5. GENERATE → 5 directions, each with a single-word descriptor tag
6. REVIEW → contact-sheet the 5 outputs, get client pick
7. LOCK → winning direction becomes clients/<slug>/brand.md for all future workRules:
- Never skip step 2. Without , every future generation for this client re-litigates brand from scratch.
brand-system.json - Always estimate before generating. 5 variants at ≈ 10 credits, but confirm in-session.
recraftv4 - Keep pitch outputs watermark-free only if the client has signed. For cold pitches, add language in prompt.
"obvious mockup" - Never mix clients in a single Drive folder or manifest. Namespace everything under .
clients/<slug>/
1. 采集 → 读取网址/演示文稿/Figma文件，提取品牌信号
2. 起草 → 生成brand-system.json提案（配色、字体风格、图像、调性、规避项）
3. 确认 → 用户确认品牌体系；保存至clients/<简称>/brand-system.json
4. 估算 → 5组方向批量生成的AI成本（目标低于1美元）
5. 生成 → 5组方向，每组配一个单字描述标签
6. 评审 → 将5组输出整理成对比图，获取客户选择
7. 锁定 → 胜出方向存入clients/<简称>/brand.md，供后续所有工作使用规则：
- 绝不能跳过步骤2。没有，后续为该客户生成的所有内容都需要重新梳理品牌定位。
brand-system.json - 生成前务必估算成本。使用生成5种变体约需10个积分，但需在会话中确认。
recraftv4 - 仅当客户已签约时，提案输出才无需加水印。对于陌生客户的提案，需在提示词中加入“明显的样稿”相关表述。
- 绝不能在单个云端文件夹或清单中混合多个客户的内容。所有内容都需归类在目录下。
clients/<简称>/
Quick Reference
快速参考
json
{
"client_slug": "acme-fintech",
"brand_system": "clients/acme-fintech/brand-system.json",
"defaults": { "model": "recraftv4", "aspectRatio": "16:9" },
"variants": {
"direction": ["editorial", "bold", "minimal", "playful", "cinematic"]
},
"jobs": [
{ "id": "{direction}", "prompt": "<client descriptor> — {direction} direction" }
]
}Lock the winning direction back into — a 10-line markdown file Claude + downstream batches can include in prompts.
clients/<slug>/brand.mdjson
{
"client_slug": "acme-fintech",
"brand_system": "clients/acme-fintech/brand-system.json",
"defaults": { "model": "recraftv4", "aspectRatio": "16:9" },
"variants": {
"direction": ["editorial", "bold", "minimal", "playful", "cinematic"]
},
"jobs": [
{ "id": "{direction}", "prompt": "<客户描述> — {direction} direction" }
]
}将胜出方向存入——这是一个10行的Markdown文件，Claude及后续批量任务可在提示词中引用。
clients/<简称>/brand.mdQuick Reference
快速参考
| Sub-task | Model | Why |
|---|---|---|
| Direction exploration (default) | | Design-forward, honors palette + type feel, cheap |
| Photo-led brands (fashion, hospitality) | | Photoreal, better lighting, for brands that live in photography |
| Typography-centric directions (editorial posters, quote slides) | | Only model that renders readable headlines reliably |
| Quick cheap drafts when testing 10+ directions | | ~1 credit each; use for throwaway iteration |
Check live IDs with before committing — names shift.
gen-ai models --mode image| 子任务 | 模型 | 原因 |
|---|---|---|
| 方向探索（默认） | | 设计导向，贴合配色+字体风格，成本低 |
| 以照片为主的品牌（时尚、酒店） | | 照片级写实，光线效果更佳，适用于依赖摄影的品牌 |
| 以字体为核心的方向（社论海报、引语幻灯片） | | 唯一能可靠渲染清晰标题的模型 |
| 测试10+方向时的快速低成本草稿 | | 每个约1积分；用于快速迭代的废弃草稿 |
提交前请通过检查实时模型ID——名称可能会变动。
gen-ai models --mode imageProcedure
操作规范
- Always scope brand before generating production work. One hour of scoping saves a week of misaligned deliverables.
- Descriptor-per-direction beats vague prompts. "Editorial" + specific cues > "something clean and modern".
- Never leak other clients' assets into a new client's prompt. No from a different client folder, ever.
-i - Save to the repo, not Drive. It's versioned alongside code and MRs.
brand-system.json - Propose 5, not 10. Client decision fatigue is real; 5 distinct directions forces a real pick.
- For NDA clients, don't name them in prompts. Use a generic descriptor ("a premium fintech for seed-stage founders") — prompts are logged.
- Build a reusable prompt library per client at — hero, tile, social, OG templates that all reference the locked direction.
clients/<slug>/prompts/
- 生成生产资源前务必先梳理品牌范围。1小时的范围梳理可避免一周的交付物偏差。
- 每个方向配专属描述词优于模糊提示词。“社论风格”+具体线索 > “简洁现代的风格”。
- 绝不能将其他客户的资产混入新客户的提示词中。绝不能使用来自其他客户文件夹的参数。
-i - 将brand-system.json保存至代码仓库，而非云端。它会与代码和MR（合并请求）一起进行版本控制。
- 提出5种方向，而非10种。客户会有决策疲劳；5种独特方向能促使客户做出明确选择。
- 对于签署NDA的客户，不要在提示词中提及他们的名称。使用通用描述（如“面向种子阶段创始人的高端金融科技品牌”）——提示词会被记录。
- 为每个客户构建可复用的提示词库，存于——包括主视觉、卡片、社交平台、OG模板，所有模板都引用已锁定的品牌方向。
clients/<简称>/prompts/
Pitfalls
常见陷阱
- Generating before locking → 5 gorgeous directions that don't fit the brand
brand-system.json - Brand cross-contamination — reusing a prior client's "because it looked nice" — your competitive moat is per-client rigor, don't blow it
brand.md - Descriptor overlap — "bold" and "cinematic" blur if prompts aren't distinct enough; each direction needs a clearly different visual hypothesis
- Over-polished scoping — if the scoping output looks like a finished ad, the client will ask "why not use this?" and you've skipped production
- Missing competitor check — landing on a direction that looks identical to the client's biggest rival
- Public Drive folder for an NDA client — always use or a private workspace
--drive-folder "internal-$CLIENT"
- 未锁定brand-system.json就开始生成 → 5组精美的方向但不符合品牌调性
- 品牌风格交叉污染 —— 复用之前客户的“因为看起来不错”——你的核心竞争力是针对每个客户的严谨性，不要破坏它
brand.md - 描述词重叠 —— 如果提示词不够明确，“大胆”和“电影感”会混淆；每个方向都需要清晰不同的视觉假设
- 过度打磨探索成果 —— 如果探索输出看起来像成品广告，客户会问“为什么不用这个？”，导致跳过生产阶段
- 遗漏竞品检查 —— 产出的方向与客户最大竞品的风格完全一致
- 为签署NDA的客户使用公共云端文件夹 —— 务必使用或私有工作区
--drive-folder "internal-$CLIENT"
Verification
验证步骤
Run to confirm authentication, then re-run the failed command with .
gen-ai whoami--debug运行确认身份验证，然后添加参数重新运行失败的命令。
gen-ai whoami--debugStep 1: Ingest + propose brand-system.json
步骤1：采集并提交brand-system.json提案
Pull signals from whatever the client gave you. If it's a URL, fetch it. If it's a deck, read it. Summarize into:
json
{
"client_slug": "acme-fintech",
"brand": {
"palette": { "primary": "#0B1F3A", "accent": "#00D4A3", "neutral": "#F5F3EE" },
"typography_feel": "geometric sans, tight tracking, editorial weights",
"imagery_direction": "abstract finance, muted photography, no stock clichés",
"tone_words": ["trustworthy", "calm", "precise"],
"do_not": ["cartoon illustration", "neon gradients", "tech-bro stock photos"]
},
"competitors": ["wise.com", "mercury.com"],
"audience": "seed-stage founders, CFO buyers",
"confidentiality": "NDA"
}Save to . This file is the single source of truth referenced by every downstream skill.
clients/<slug>/brand-system.json从客户提供的资料中提取品牌信号。如果是网址，获取其内容；如果是演示文稿，读取内容。整理为如下格式：
json
{
"client_slug": "acme-fintech",
"brand": {
"palette": { "primary": "#0B1F3A", "accent": "#00D4A3", "neutral": "#F5F3EE" },
"typography_feel": "geometric sans, tight tracking, editorial weights",
"imagery_direction": "abstract finance, muted photography, no stock clichés",
"tone_words": ["trustworthy", "calm", "precise"],
"do_not": ["cartoon illustration", "neon gradients", "tech-bro stock photos"]
},
"competitors": ["wise.com", "mercury.com"],
"audience": "seed-stage founders, CFO buyers",
"confidentiality": "NDA"
}保存至。该文件是所有后续任务引用的唯一可信来源。
clients/<简称>/brand-system.jsonStep 2: Generate 5 directions
步骤2：生成5组方向
Each direction gets a single-word descriptor. Standard starter set: editorial, bold, minimal, playful, cinematic. Swap any that clashes with the brand's tone words (e.g. drop "playful" for a private bank).
bash
CLIENT="acme-fintech"
OUT="clients/$CLIENT/scoping"
mkdir -p "$OUT"每组方向配一个单字描述词。标准初始集合：editorial（社论风）、bold（大胆风）、minimal（极简风）、playful（趣味风）、cinematic（电影感）。如果某个词与品牌调性冲突，可替换（例如，为私人银行移除“playful”）。
bash
CLIENT="acme-fintech"
OUT="clients/$CLIENT/scoping"
mkdir -p "$OUT"Check per-call pricing first, then multiply by 5 directions.
先检查单次调用的价格，再乘以5组方向。
gen-ai pricing recraftv4
gen-ai pricing recraftv4
Run as 5 discrete jobs so each can get its own direction prompt
拆分为5个独立任务，以便为每组方向配置专属提示词
cat > /tmp/$CLIENT-scoping.json <<EOF
{
"defaults": { "model": "recraftv4" },
"jobs": [
{ "id": "01-editorial", "prompt": "$CLIENT hero — editorial magazine feel, restrained palette, generous whitespace, confident typography", "aspectRatio": "16:9" },
{ "id": "02-bold", "prompt": "$CLIENT hero — bold graphic, high contrast, oversized type, flat color blocking", "aspectRatio": "16:9" },
{ "id": "03-minimal", "prompt": "$CLIENT hero — minimal composition, single focal element, muted neutrals, precise negative space", "aspectRatio": "16:9" },
{ "id": "04-playful", "prompt": "$CLIENT hero — playful abstract shapes, warm palette, hand-drawn accent, approachable energy", "aspectRatio": "16:9" },
{ "id": "05-cinematic", "prompt": "$CLIENT hero — cinematic photography, shallow depth of field, moody lighting, filmic grain", "aspectRatio": "16:9" }
]
}
EOF
gen-ai batch run /tmp/$CLIENT-scoping.json -c 5 -o "$OUT"
Results land at `clients/<slug>/scoping/0X-<direction>.webp` + `results.json`. Present the 5 side-by-side to the client.cat > /tmp/$CLIENT-scoping.json <<EOF
{
"defaults": { "model": "recraftv4" },
"jobs": [
{ "id": "01-editorial", "prompt": "$CLIENT hero — editorial magazine feel, restrained palette, generous whitespace, confident typography", "aspectRatio": "16:9" },
{ "id": "02-bold", "prompt": "$CLIENT hero — bold graphic, high contrast, oversized type, flat color blocking", "aspectRatio": "16:9" },
{ "id": "03-minimal", "prompt": "$CLIENT hero — minimal composition, single focal element, muted neutrals, precise negative space", "aspectRatio": "16:9" },
{ "id": "04-playful", "prompt": "$CLIENT hero — playful abstract shapes, warm palette, hand-drawn accent, approachable energy", "aspectRatio": "16:9" },
{ "id": "05-cinematic", "prompt": "$CLIENT hero — cinematic photography, shallow depth of field, moody lighting, filmic grain", "aspectRatio": "16:9" }
]
}
EOF
gen-ai batch run /tmp/$CLIENT-scoping.json -c 5 -o "$OUT"
结果将保存至`clients/<简称>/scoping/0X-<direction>.webp` + `results.json`。将5组输出并排展示给客户。Cost & time
成本与时间
| Phase | Typical spend | Typical time |
|---|---|---|
Scoping (this skill, 5 directions, | ~$0.50–$1 | 15-30 min incl. review |
| Production (after direction lock) | $5–$50 per asset set | Per deliverable |
Keep scoping under $1. If a client won't approve a direction after 5, the brief is broken — don't burn credits, push back on the brief.
| 阶段 | 典型花费 | 典型耗时 |
|---|---|---|
范围界定（本流程，5组方向，使用 | ~0.5–1美元 | 15-30分钟（含评审） |
| 生产阶段（锁定方向后） | 每组资产5–50美元 | 视交付物而定 |
确保范围界定成本低于1美元。如果客户在5组方向后仍无法确认，说明brief存在问题——不要浪费积分，需重新梳理brief。
See also
相关链接
- — once direction is locked, produce pitch assets
workflows/agency-pitch-mockups/ - — run scoping-plus-production across many retainer clients
workflows/agency-multi-brand-pack/ - — package final work at engagement end
workflows/agency-client-handoff/ - — general multi-step patterns
gen-ai-workflows.md - — manifest shapes and concurrency tuning
gen-ai-batch.md
- —— 锁定方向后，生成提案资产
workflows/agency-pitch-mockups/ - —— 为多个固定合作客户执行范围界定+生产流程
workflows/agency-multi-brand-pack/ - —— 项目结束时打包最终成果
workflows/agency-client-handoff/ - —— 通用多步骤模式
gen-ai-workflows.md - —— 清单格式和并发调优
gen-ai-batch.md