gpt-image-gen

Featured

Use when the user asks to generate an image via GPT/Codex (e.g. 「叫 gpt 生圖」「幫我用 gpt 生圖」「gpt 畫一個 X」). The skill drafts a Chinese + English prompt pair, iterates with the user until they explicitly approve, then dispatches Codex CLI ($imagegen skill, codex built-in image_gen) in the background, monitors progress, converts the result to a jpg in the current working directory, and writes a sidecar prompt log. Does text-to-image AND img2img — drop a reference image (on-disk file) and it runs Codex `-i` to lock a face/character across scenes.

AI & Automation 76 stars 13 forks Updated 6 days ago MIT

Install

View on GitHub

Quality Score: 92/100

Stars 20%
63
Recency 20%
100
Frontmatter 20%
70
Documentation 15%
100
Issue Health 10%
50
License 10%
100
Description 5%
100

Skill Content

# gpt-image-gen — 用 Codex CLI 叫內建 image_gen 生圖 You are a prompt-crafting partner who turns the user's loose Chinese description into a tight bilingual prompt pair, iterates with the user until they explicitly approve, then dispatches Codex CLI to generate the image. You are **not** the image generator — Codex is. Your job is prompt design, user confirmation gating, and execution orchestration. **CRITICAL — 三條紅線**: 1. **未拍板絕不呼叫 codex** — 拍板 = user 明確說 `OK` / `生` / `go` / `下去`。其他正向回應(「不錯」「可以喔」「應該行」)一律當「還沒拍板」處理,繼續等明確指令。生圖會花 user 的錢,誤觸發 = 違規。 2. **有 reference image → 走 img2img**(codex `-i`) — user 這輪有附底圖(拖曳/貼上/`[Image #N]`)→ Step 3a 偵測 → Step 4 用 `codex exec ... -i <ref>` 跑 img2img(鎖臉/角色一致)。⚠️ codex `-i` 吃**本機檔案路徑**:底圖有實體檔就 img2img;只貼在對話、本機無檔 → 問 user 要路徑,給不出才退回印 prompt 貼 GUI。**拍板 gate(紅線 1)對 img2img 一樣適用。** 3. **不寫死任何預設風格** — Skill 不存 style preset。每張圖風格純靠當下 conversation context + user 描述推。沒 context 就問。 --- ## Step 1: 判斷觸發語境(mid-conversation vs 新對話) 讀 trigger 那輪訊息 + 最近 5-10 輪 context,落到下表: | 情況 | 動作 | |------|------| | Mid-conversation 且 context 含 ≥3 錨點(**場景 + 主體 + 動作**) | 跳 Step 2,直接展 prompt | | Mid-conversation 但 context 不足(缺任一錨點) | 跳 Step 1a,互動補問 | | 新對話 / 純 trigger 沒帶任何描述 | 跳 Step 1a,互動問清楚 | **錨點判斷標準**: - **場景**:哪裡 / 什麼背景(街道、室內、特定地標、純色底...) - **主體**:誰 / 什麼物件、外型描述 - **動作**:在做什麼 / 姿態 / 表情 / 互動 不確定時就降級到 Step 1a 問清楚 — **禁止靠想像力填空**。 ### Step 1a: 互動補問(只在缺錨點時) 問題收斂在缺的那幾項,每次最多 3 個問題、numbered list、口語: ``` 要先確認幾件事再展 prompt: 1. 場景 / 背景:__ 2. 主體:誰 / 什麼,幾個,長相 / 外型描述:__ 3. 動作 /...

Details

Author
KerberosClaw
Repository
KerberosClaw/kc_ai_skills
Created
4 months ago
Last Updated
6 days ago
Language
Python
License
MIT

Integrates with

Similar Skills

Semantically similar based on skill content — not just same category

AI & Automation Listed

prompt-coach

Turns a vague, underspecified request into a clearly structured Goal / Constraints / Obstacle breakdown, then produces a ready-to-use structured prompt. Use this whenever the user brings a fuzzy ask ("幫我想辦法...", "有什麼方法可以...", "我不知道怎麼處理..."), asks explicitly to help them "問對問題" or structure their request, or describes a real-world problem/decision they want AI help with but haven't yet specified what success looks like, what can't change, or what the actual blocker is. Trigger this proactively — even without the user naming the skill — any time answering directly would likely produce a generic, "technically correct but not actually usable" answer rather than something tailored to their real situation.

0 Updated 1 weeks ago
goingli0324
AI & Automation Listed

prompt-coach

Turns a vague ask into a Goal / Constraints / Obstacle breakdown, then a ready-to-use structured prompt. Use when the user brings a fuzzy request (「幫我想辦法…」「有什麼方法可以…」「我不知道怎麼處理…」), asks to 「問對問題」, or names a real problem without saying what success looks like, what can't change, or what the blocker is.

0 Updated 2 days ago
goingli0324
AI & Automation Listed

parallel-imagegen

Run multiple built-in Codex image_gen calls concurrently by assigning each output to an independent codex exec process, with up to 12 active workers, isolated retries, and thread-evidence verification. Use for two or more independent raster generation or editing outputs, even when prompts, references, subjects, or styles differ. Do not use for one output, unresolved sequential dependencies, or an explicitly selected Images API / imagegen CLI workflow. 通过为每个输出启动独立 Codex 进程,并发生成或编辑两个 及以上互不依赖的位图;支持最多 12 个 Worker、失败隔离重试和线程证据验证。 不用于单张输出、尚未解决的前后依赖,或用户明确选择 Images API / imagegen CLI 的场景。

3 Updated 4 days ago
Bing-Bryan