ad-campaign

Compare original and translation side by side

🇺🇸

Original

English

🇨🇳

Translation

Chinese

Ad Campaign — plan → generate → judge → package

广告活动 — 规划→生成→评估→打包

A repeatable pipeline for shipping a set of on-brand image ads at volume. Built from a real 30-ad campaign that went 75/75 accepted at ~7.5 credits per shipped asset, then re-validated cold on a smaller executor model with zero improvisation.

Core loop: copy is judged as text before it costs credits as pixels; every image is judged with eyes before it's called done.

一套可重复的工作流，用于批量产出符合品牌调性的图片广告。基于一个真实的30条广告活动构建，该活动75条广告全部通过审核，每条产出素材约花费7.5 credits，之后在更小的执行模型上重新验证，全程无即兴发挥。

核心循环：文案先以文本形式评估，再花费credits生成像素内容；每张图片在定稿前都需人工视觉评估。

Design note — no improvisation required

设计说明 — 无需即兴发挥

The creative judgment is front-loaded into frozen assets. Execute them, don't re-derive them:

Fixed prompt blocks are FROZEN STRINGS (
```
references/prompt-template.md
```
). Fill the brand slots once, then only the four per-ad slots ({SCENE}/{hook}/{key}/{support}) change. Never paraphrase, "improve" or re-order the fixed clauses — identical wording across the set is what makes 30 ads read as one campaign.
Scenes are built by formula, then remixed from your accepted-scene library (see "Grow the library" below) — not invented from zero each time.
Judging is literal rubric application (
```
references/judge-rubric.md
```
), not vibes. Uncertain between FLAG and FAIL → FLAG and log. The ignore-list is binding.
Decision points are enumerated in Step 0. Anything outside them: reversible → make the obvious call and log it; spends meaningfully beyond the agreed budget or changes brand/copy rules → stop and ask the user.

创意判断已预先固化到固定资产中。只需执行，无需重新推导：

固定提示词块为冻结字符串（
```
references/prompt-template.md
```
）。只需填充一次品牌占位符，之后仅需更改每条广告对应的四个占位符（{SCENE}/{hook}/{key}/{support}）。绝不要改写、“优化”或调整固定条款的顺序——整套广告使用完全一致的表述，才能让30条广告看起来属于同一活动。
场景按公式构建，再从已通过审核的场景库中混合生成（见下文“扩充库内容”）——不要每次从零开始创建。
评估需严格按照评分标准执行（
```
references/judge-rubric.md
```
），而非凭感觉。若在标记FLAG和FAIL之间犹豫→标记FLAG并记录日志。忽略列表具有约束力。
决策点已在步骤0中列出。超出范围的情况：可逆转→做出明确决策并记录；若支出明显超出约定预算或更改品牌/文案规则→停止并询问用户。

When to activate

启用时机

"build an ad campaign / a set of ads / paid-social creative"
A batch of on-brand image ads at volume (10+), with text on-image
Re-running the system for a new brand (swap the brand kit, keep the pipeline)

请求“打造广告活动/一组广告/付费社交创意”时
需要批量（10条以上）带图上文案、符合品牌调性的图片广告时
为新品牌重新运行该系统时（替换品牌套件，保留工作流）

Step 0 — Lock the variables (intake)

步骤0 — 锁定变量（需求收集）

Ask the user; if they're unavailable, lock the recommended defaults, state them as veto-able, and keep moving.

Variable	Recommended default
Angles	The brand's active sales angles/offers, from wherever the brand keeps them. A NEW angle gets written into the brand doc FIRST, then used — brand context outlives the campaign.
Volume	30 masters = 1/day for a month. Direction gate of 2–3 before the rest.
Placements	Meta feed + story. gpt_image_2 has NO 4:5 → generate 3:4 composed to survive a TOP-trim crop (headline upper-third, CTA/footer at the bottom edge).
Story twins	Default OFF. If feed posts get reposted to stories anyway, 9:16 twins double the credit spend for the same creative. If a one-off vertical is explicitly requested, re-stage it (see template), never resize.
Sequences	One 3-frame 9:16 arc per angle (distinct creative, not a twin): hook (no button) → turn (no button) → CTA (prominent pill). Button only on f3 is what makes it an arc.
CTA register	ADS ARE ASSERTIVE: solid accent-colored {CTA_TEXT} pill + cost-of-inaction urgency. Soft asks ("worth a quick chat?") belong in email — never on paid placements.
Audience	BROAD + diverse visual worlds, enforced by quota (Step 2). Never let one niche dominate.
Footer/domain	{FOOTER_TEXT} tiny mono at the extreme bottom edge.
Captions	Full pack per ad: hook as first line, founder-voice body, CTA → domain, 4–6 real hashtags.
Budget	7 credits/image at quality high + 2k. Check `higgsfield account status` before and after.

询问用户；若用户无法回应，锁定推荐默认值，说明这些值可被否决，然后继续推进。

变量	推荐默认值
营销角度	品牌当前的销售角度/优惠信息，取自品牌存储该内容的渠道。新角度需先写入品牌文档，再投入使用——品牌上下文的生命周期长于单个广告活动。
数量	30条主广告=每月1条。批量生成前先进行2–3条的方向审核。
投放位置	Meta信息流+故事。gpt_image_2不支持4:5比例→生成3:4比例的素材，确保裁剪顶部后仍能正常展示（标题位于上三分之一，CTA/页脚位于底部边缘）。
故事版适配	默认关闭。若信息流帖子会被重新发布到故事版，生成9:16的适配版本会使credits花费翻倍，但创意内容相同。若明确要求单独的竖版素材，需重新制作（参考模板），绝不要直接调整尺寸。
故事序列	每个营销角度对应一个9:16比例的三帧故事弧：钩子（无按钮）→转折（无按钮）→CTA（显眼按钮）。仅在第三帧添加按钮，才构成完整的故事弧。
CTA风格	广告需态度明确：纯色强调色的{CTA_TEXT}按钮+不作为的代价类紧迫感。柔和请求（“值得快速聊聊吗？”）适合邮件场景——绝不要用于付费投放位置。
受众	广泛+多样化视觉风格，通过配额强制实现（步骤2）。绝不要让单一细分领域占据主导。
页脚/域名	{FOOTER_TEXT}以小号等宽字体放在最底部边缘。
文案	每条广告配套完整文案：首行是钩子，创始人语气的正文，CTA→域名，4–6个真实话题标签。
预算	高质量+2k分辨率的图片每条花费7 credits。生成前后需检查 `higgsfield account status` 。

Step 1 — Fill the brand kit (once per brand)

步骤1 — 填充品牌套件（每个品牌仅需一次）

These slot into the frozen template and then get treated as frozen themselves:

Slot	What it is	Example
`{BG_HEX}`	Deep near-black base grade	`#0A0A0B`
`{TEXT_HEX}`	Off-white text color	`#F4F4F0`
`{ACCENT_HEX}`	The ONE accent color	`#E4572E`
`{CTA_TEXT}`	Ad CTA label, lowercase	`book a call`
`{FOOTER_TEXT}`	Brand — domain footer line	`ACME — acmestudio.com`

Read the brand's positioning/ICP/voice docs if they exist; the pipeline is brand-agnostic, the taste bar isn't.
Exactly ONE accent color — restraint is what makes a set read premium.
The template's type classes (heavy lowercase grotesque headline + monospace support) are part of the proven recipe. Swap them for the brand's own type only if you must — and re-run the direction gate if you do.

这些内容会填入冻结模板，之后也会被视为冻结内容：

占位符	说明	示例
`{BG_HEX}`	深近黑色底色	`#0A0A0B`
`{TEXT_HEX}`	米白色文本颜色	`#F4F4F0`
`{ACCENT_HEX}`	唯一强调色	`#E4572E`
`{CTA_TEXT}`	广告CTA标签，小写	`book a call`
`{FOOTER_TEXT}`	品牌——域名页脚文字	`ACME — acmestudio.com`

若品牌有定位/理想客户画像/语气文档，请阅读；工作流是品牌无关的，但审美标准不是。
仅使用一种强调色——克制才能让整套广告看起来高端。
模板中的字体类别（粗体无衬线小写标题+等宽辅助文字）是经过验证的方案。仅在必要时替换为品牌自有字体——若替换，需重新进行方向审核。

Step 2 — Campaign plan + copy matrix (before any prompt)

步骤2 — 活动规划+文案矩阵（生成提示词前）

Write

plan/campaign-plan.md

plan/copy-matrix.md

. Every ad exists as a row first: id · angle · hook · key · support · world.

Copy rules (voice-checked as text, cheap):

Hook ≤9 words, lowercase, blunt. Support ≤12 words. No exclamation marks, no AI-tell words, no fake urgency (cost-of-inaction urgency OK).
Exactly ONE keyword per hook in the accent color.
Never name the niche out loud — the visual world codes who-it's-for (a docket rail says "restaurant" without saying it). World-cue NOUNS inside the pain are fine ("service ends at ten"); niche LABELS are not ("for restaurant owners").
No client names. Sell WHAT it does / the outcome, never HOW / the internal mechanics. No domain in headlines.
World-spread quota, written down before generating: across a 30-ad set, no world more than ~3×, cap your most-tempting world, include several universal metaphors and at most 2 type-only ads. For a sub-batch apply the spirit: every world distinct, no repeats of recent campaigns' worlds.
Mix day and night scenes across the set; grime is banned — premium-worn yes (patina, aged brass, worn oak), filthy never (dirt/mud/stains/crusted = judge FAIL). A dark grade holds in daylight too.

撰写

plan/campaign-plan.md

plan/copy-matrix.md

。每条广告先以行形式存在：id · 营销角度 · 钩子 · 核心信息 · 辅助信息 · 视觉风格。

文案规则（以文本形式检查语气，成本低）：

钩子≤9个单词，小写，直白。辅助信息≤12个单词。无感叹号，无AI生成痕迹的词汇，无虚假紧迫感（不作为的代价类紧迫感可接受）。
每个钩子中恰好有一个关键词使用强调色。
绝不要直接说出细分领域——视觉风格会暗示目标受众（比如案卷轨道代表“餐厅”，无需明说）。痛点中的场景提示名词可接受（“服务10点结束”）；但细分领域标签不可接受（“面向餐厅老板”）。
不要提及客户名称。推销产品功能/结果，绝不要推销实现方式/内部机制。标题中不要出现域名。
提前记录视觉风格分布配额：30条广告的套装中，同一种视觉风格出现不超过~3次，限制最常用的风格，加入几个通用隐喻，最多包含2条纯文字广告。对于子批次，遵循此原则：每种视觉风格都不同，不要重复近期活动的风格。
套装中混合日景和夜景；禁止脏乱元素——允许有使用痕迹的质感（铜绿、旧黄铜、磨损橡木），但绝对禁止肮脏元素（灰尘/泥土/污渍/结痂=评估FAIL）。深色底色在日光下也能保持效果。

Step 3 — Build prompts from the template

步骤3 — 从模板构建提示词

One

.txt

per ad:

prompts/final/{id}.txt

(3:4 masters),

prompts/seq/{angle}seq-f{1,2,3}.txt

(9:16 sequences). The full template with clause-by-clause reasoning:

references/prompt-template.md

. Non-negotiable clauses: all on-image text quoted + "exactly as quoted"; any incidental surface text forced to "illegible blur"; ends with "No other readable text anywhere, no watermark, no logo, no border."

每条广告对应一个

.txt

文件：

prompts/final/{id}.txt

（3:4主广告），

prompts/seq/{angle}seq-f{1,2,3}.txt

（9:16序列）。带有逐条理由的完整模板：

references/prompt-template.md

。不可协商的条款：所有图上文案需加引号并标注“严格按照引号内内容”；任何附带的表面文本需强制设置为“模糊不可读”；结尾需标注“无其他可读文本，无水印，无logo，无边框”。

Step 4 — Direction gate (before spending the batch)

步骤4 — 方向审核（批量生成前）

Render 2–3 covering the distinct visual modes (one human scene, one macro metaphor, one type-only). Judge them (Step 6 rubric). Only a pass on all modes unlocks the full batch. If the user is present, show them the gate renders.

渲染2–3条覆盖不同视觉模式的广告（一个人物场景，一个宏观隐喻，一个纯文字）。按照步骤6的评分标准评估。只有所有模式都通过审核，才能解锁完整批量生成。若用户在场，展示审核用的渲染图。

Step 5 — Generate

步骤5 — 生成

bash

higgsfield generate create gpt_image_2 \
  --prompt "$(cat prompts/final/a1.txt)" \
  --aspect_ratio 3:4 --quality high --resolution 2k \
  --wait --json > generated/a1-r1.json

Full-wave runs: batch sequentially inside a background task. Tight-cap runs (dry runs, gates, ≤5 gens): serialize in the foreground and inspect each JSON before firing the next call — a failed generation still counts against a hard cap.
Extract
```
[0]['result_url']
```
from each JSON → curl to
```
generated/{id}-{ratio}-r{round}.png
```
. Inspect the JSON before curling — a 502 writes HTML into the file, and an empty result_url means the render failed server-side.
Auth dies occasionally:
```
higgsfield auth login
```
(device flow) — surface the URL to the user (copy it to their clipboard if you can). Balance:
```
higgsfield account status
```
.

bash

higgsfield generate create gpt_image_2 \
  --prompt "$(cat prompts/final/a1.txt)" \
  --aspect_ratio 3:4 --quality high --resolution 2k \
  --wait --json > generated/a1-r1.json

全量运行：在后台任务中按顺序批量生成。限量运行（试运行、审核、≤5条生成）：在前台序列化执行，每生成一个JSON就检查一次，再执行下一个请求——生成失败仍会占用硬上限额度。
从每个JSON中提取
```
[0]['result_url']
```
→通过curl保存到
```
generated/{id}-{ratio}-r{round}.png
```
。在curl前务必检查JSON——502错误会将HTML写入文件，空的result_url表示服务器端渲染失败。
认证偶尔会失效：执行
```
higgsfield auth login
```
（设备流）——将URL展示给用户（若可以，复制到用户剪贴板）。余额查询：
```
higgsfield account status
```
。

Step 6 — Judge loop (eyes on every image)

步骤6 — 评估循环（每张图片都要人工查看）

Read every PNG with vision. Score 7 criteria: text accuracy · instant read · who-is-this-for · no slop · brand world · slap factor · craft. Verdict per image with reasons →

judgement-log.md

. Full rubric, thresholds, and the known-nits ignore list:

references/judge-rubric.md

Accept = zero FAIL and ≤1 FLAG. Rejects regen ≤2 rounds, then the concept is cut, not forced.
Credit discipline: set-level nits invisible at delivery size (apostrophe glyphs, pill drift, footer under platform chrome) are logged, never regenned.
Known failure modes: 502 → re-run same command; content filter (empty result_url twice) → de-brand the object noun ("whiskey cask" → "oak barrel aging cellar").

用视觉检查每张PNG图片。评估7项标准：文本准确性 · 即时可读性 · 目标受众清晰度 · 无瑕疵 · 品牌风格匹配 · 吸引力 · 制作质量。为每张图片给出评估结论及理由→写入

judgement-log.md

。完整评分标准、阈值、已知问题忽略列表：

references/judge-rubric.md

。

通过=0项FAIL且≤1项FLAG。被拒绝的素材最多重新生成2轮，之后放弃该创意，不要强行推进。
credits管控：交付尺寸下不可见的套装级小问题（撇号字形、按钮偏移、页脚被平台栏遮挡）只需记录，无需重新生成。
已知失败模式：502错误→重新运行相同命令；内容过滤（连续两次空result_url）→移除品牌相关的名词（“威士忌桶”→“橡木桶陈酿窖”）。

Step 7 — File + package

步骤7 — 文件整理+打包

Accepted →
```
accepted/
```
(flat archive), then
```
final/
```
per-angle folders: masters +
```
sequence/
```
+ that angle's
```
captions.md
```
+ top-level
```
README.md
```
(cadence, ad-button choice, crop rules, honest flags, provenance).
Captions: hook as line 1 (captions truncate), founder-voice body,
```
{CTA_TEXT} → domain
```
, 4–6 hashtags. Meta button: Contact Us / Get Quote.

通过审核的素材→存入
```
accepted/
```
（扁平归档），之后按营销角度存入
```
final/
```
文件夹：主广告+
```
sequence/
```
+对应角度的
```
captions.md
```
+顶层
```
README.md
```
（发布节奏、广告按钮选择、裁剪规则、明确标记的问题、来源说明）。
文案：首行是钩子（文案会被截断），创始人语气的正文，
```
{CTA_TEXT} → 域名
```
，4–6个话题标签。Meta平台按钮：联系我们/获取报价。

Step 8 — Honest report

步骤8 — 如实报告

Accepted/rejected counts with reasons, every flag surfaced (never buried), credit spend before/after, paths. Give paths — don't auto-open files.

Running as a SUBAGENT: harness guardrails may block writing a REPORT.md file — return the full report as your final text instead. The on-disk artifacts (plan/copy-matrix.md, judgement-log.md, prompts/, generated/) write fine and are the audit trail.

包含通过/拒绝数量及原因，所有标记的问题都要列出（绝不要隐藏），生成前后的credits花费，操作路径。提供路径——不要自动打开文件。

作为SUBAGENT运行时：防护规则可能会阻止写入REPORT.md文件→将完整报告作为最终文本返回。磁盘上的 artifacts（plan/copy-matrix.md, judgement-log.md, prompts/, generated/）可正常写入，作为审计追踪记录。

Grow the library

扩充库内容

Keep every accepted ad's

{SCENE}

sentence and its verdict from

judgement-log.md

. That corpus is the next campaign's scene library — remix proven scenes instead of inventing from zero, and the judge log becomes training data for what your brand accepts.

保存每条通过审核的广告的

{SCENE}

语句及其在

judgement-log.md

中的评估结论。该语料库将作为下一次广告活动的场景库——混合已验证的场景，而非从零开始创建，评估日志也会成为品牌接受标准的训练数据。

References

参考文件

File	Contents
`references/prompt-template.md`	Master template + sequence rules, clause-by-clause WHY
`references/judge-rubric.md`	7 criteria, accept thresholds, regen policy, ignore-list, filter/502 fixes

文件	内容
`references/prompt-template.md`	主模板+序列规则，逐条说明设计理由
`references/judge-rubric.md`	7项评估标准、通过阈值、重新生成规则、忽略列表、过滤/502错误修复方案