What is visual-art-direction?

A visual-direction skill for improving ordinary portraits, selfies, and photos of two people while preserving a natural appearance. It turns the user's intended feeling and use into an editing strategy and checks the result.

Use it when preparing everyday portraits for social posts such as Xiaohongshu, comparing a small number of image options, or making further adjustments based on feedback.

It helps decide whether a portrait needs cropping, lighting, color, or restrained local adjustments without assuming that a vague request means retouching, reshaping, or changing the background.

visual-art-direction — Skill for Claude Code and Codex by weibinyuvert

Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

Any agent

npx skills add weibinyuvertex120/seeform --skill visual-art-direction

Clone the repo

git clone --depth 1 https://github.com/weibinyuvertex120/seeform

Made for: Claude Code, Codex.

Wrote this? Show the measurements

A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.

README.md

[![agentmods](https://agentmods.dev/badge/skills/weibinyuvertex120/seeform/visual-art-direction/github.svg)](https://agentmods.dev/skills/weibinyuvertex120/seeform/visual-art-direction)

Your own site

<a href="https://agentmods.dev/skills/weibinyuvertex120/seeform/visual-art-direction"><img src="https://agentmods.dev/badge/skills/weibinyuvertex120/seeform/visual-art-direction/github.svg" alt="Measured on agentmods" height="20"></a>

Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.

Your own site · 80×15

<a href="https://agentmods.dev/skills/weibinyuvertex120/seeform/visual-art-direction"><img src="https://agentmods.dev/badge/skills/weibinyuvertex120/seeform/visual-art-direction.svg" alt="Reviewed on agentmods" width="80" height="20"></a>

Per session 129 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.

When invoked 3,628 The whole file, excluding the scripts and references it only reads on demand.

Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe.

Origin original No closer match found in the catalogue.

Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

Model	Per session	Once invoked
Fable 5.1	$0.00129	$0.03628
Opus 5	$0.00064	$0.01814
Sonnet 5	$0.00026	$0.00726
Haiku 4.5	$0.00013	$0.00363

Measured 10d ago against content hash bd0eab94a7df, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-10, from the pricing page.

Security

Grade A, and why

visual-art-direction scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 10d ago.

The scan reads SKILL.md. This mod also ships 23 executable files (scripts/__init__.py, scripts/__main__.py, scripts/adapter_contract.py, …), listed below but not scanned — reading those needs a real analyzer, not pattern matching.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

skills/visual-art-direction/SKILL.md · 229 lines

How it starts

The opening of the file, as written. The whole thing — 229 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Visual Art Direction · 见相

定位与触发

见相是宿主 Agent 内部、面向普通生活人像的视觉判断与转化层。它观察素材、识别主要视觉问题和潜力，把宿主已经理解的用户意图编译成编辑器无关的视觉策略，并在执行后检查策略是否得到满足。插图、截图、视频帧、封面和概念视觉保留为未来兼容，不纳入 V1 主验收。

适合触发：

用户提供生活照、自拍、双人照或普通人像，希望更自然、更有状态、更出片或适合小红书等社交发布。
用户已经表达感觉或用途，但不确定该裁切、调光、调色或做克制的局部处理。
需要比较少量人像候选、解释真实性取舍，或根据一句反馈继续调整。

不要触发：

纯软件按钮教程、文件格式转换或已给出精确参数的机械操作。
只需生成提示词、罗列风格名词或推荐滤镜。
没有视觉素材或可靠观察能力，却要求断言画面事实。
主要目标是插图、截图、视频帧、海报或其他非生活人像媒介；这些属于未来兼容范围，不作为 V1 触发。

见相保持一个公开 Skill。内部模块是逻辑职责，不是独立 Agent、微服务或新的用户入口。

宿主、Seeform、Adapter 与用户职责

层	唯一职责
用户	表达目标与边界；接受、拒绝或继续调整；决定高风险改变
宿主 Agent	理解当前对话、任务、用途、历史约束和反馈；调用见相；收集用户决定
Seeform	结合原图事实验证上下文，形成 VisualStrategy，执行风险门并评价策略结果
Adapter	声明真实能力，在能力边界内执行已批准策略并返回工程证据

边界规则：

宿主不替视觉模型判断照片光线、构图或主体竞争。
Seeform 不要求用户重复宿主已经提供的信息，也不擅自定义用户审美。
Adapter 不把“更有状态”解释成磨皮、瘦脸、换背景或其他未批准动作。
Evaluator 不生成用户接受、拒绝或修改决定。
长期偏好由宿主管理；单次反馈只是一条偏好证据。

最小输入与直接用户输出

用户可以只给一张素材和一句话。宿主应将当前语境整理为 HostIntentContext，至少保留能影响视觉决策的内容：

{
  "user_request": "自然但更有状态，适合发小红书",
  "constraints": [
    {
      "constraint_id": "constraint-1",
      "kind": "forbid",
      "statement": "不要修白",
      "priority": "hard",
      "source": "user",
      "scope": "case"
    }
  ],
  "use_case": "social_portrait",
  "change_tolerance": "low",
  "feedback_refs": []
}

不要把内部字段变成用户问卷。旧的字符串约束可以进入兼容归一化，但必须标为 unclassified，不能按关键词猜测语义。

默认由宿主只对外交付三件事：一句视觉观点、一个默认结果，以及在高风险或能力不足时的一次简短确认/限制说明。不要默认展示 JSON、hash、P01-P10、L1-L4 推理链或能力探测。只有真实分歧才给少量候选；用户的一句话反馈应绑定当前候选并形成最小策略修订，不重新抽取无关方向。

观察到反馈的工作流

HostIntentContext
  + SourceObservation
  -> VisualStrategy
  -> Policy Gate
  -> Adapter Negotiation
  -> AdapterExecutionRequest
  -> ExecutionResult
  -> TechnicalValidation + StrategyEvaluation
  -> UserFeedback
  -> FeedbackRoutingDecision

1. 观察

只记录素材中可验证的事实和不确定性。宿主上下文用于决定观察重点和检查冲突，但不是图像证据。没有观察能力时返回 blocked_no_view_capability，不得补写画面事实。

需要系统诊断构图、光线、色彩、主体、空间、叙事和可恢复性时，读取 problem-diagnosis.md。

2. 编译视觉策略

Read the full file on GitHub · 229 lines

Files

What ships with it

35 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

10d ago First seen · 229 lines · 129 tokens per session scan A bd0eab94a7df

Subscribe to this mod's changes

visual-art-direction is a skill published in the GitHub repository weibinyuvertex120/seeform (2 stars, last pushed 16d ago), licensed Apache-2.0. It adds 129 tokens to every session and 3,628 once invoked, about $0.0006 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.

Creative and media agent-skills ai-agent computer-vision creative-tools image-editing image-understanding

watching-videos

The user shared a video URL, a YouTube/TikTok/stream link, a local video file, a screen recording, a meeting recording, or a playlist/folder of videos — "watch this", "summarize this video", "what's in this recording". Use this to actually watch the video — download, extract frames, OCR, transcribe, and index it …

oxbshw/watch-skill · 97 tokens

extracting-structure

The user wants structure pulled out of a watched video — "make chapters for this video", "where does the bug appear in this recording", "turn this screen recording into a bug report", "how strong is my intro/hook". Use this for deterministic extraction from the index — chapters with timestamps, a fileable bug report…

oxbshw/watch-skill · 81 tokens

xxg-portrait-rebuild-light

Edit an existing JPG, JPEG, PNG, or WebP portrait to rebuild physically coherent light, exposure, color, capture style, and clean optical skin realism without changing the person. Use for plastic-skin or AI-look removal, tone or white-balance correction, camera/film/device emulation, natural fill, backlight…

moskoo/xxg-portrait-rebuild-light · 101 tokens

replicate-images

Use when generating or editing images through the Replicate API, or when a modern image model keeps ignoring the prompt — choosing aspect ratio, resolution, seed and output format, image-to-image, multi-reference composition, text-driven inpainting, readable text inside a generated image, and structuring prompts per…

ericrisco/rsc-harness · 84 tokens

stable-diffusion-image-generation

State-of-the-art text-to-image generation with Stable Diffusion models via HuggingFace Diffusers. Use when generating images from text prompts, performing image-to-image translation, inpainting, or building custom diffusion pipelines.

davila7/claude-code-templates · 50 tokens

image-to-psd

A skill that reconstructs images as editable, layered Photoshop files. It separates the repaired background, visual components, and text into layers instead of leaving the page as one flat image.

DSY-Xueai/image2editable · 75 tokens