gemini-image-describer
Use this skill when a user provides an image (local file or URL) and wants a detailed Traditional Chinese description of the image content.
Skill by duotify
Install
Works with Claude Code, Claude apps, Codex, Cursor, Gemini CLI, GitHub Copilot, VS Code, OpenCode, Goose, Windsurf. Read the repository before you give any tool access to your data.
npx agentmarket install skill/duotify-gemini-image-describer
Asks which of your tools to add it to (Claude Code, Codex, Cursor, VS Code, Windsurf, Gemini CLI, Claude Desktop) and shows the change first. See https://agentmart.dev/cli.
Get the weekly diff
The weekly diff: one email a week with new agent tools, notable updates and grade changes, plus security alerts for tools you follow. Double opt-in, one-click unsubscribe.
About
Use cases: writing
Reviews
Write a reviewNo reviews yet. Installed it? Reviews come from verified installers only. How reviews work
Permissions
What this skill can do and what it asks for, from its SKILL.md and code. Checked Oct 25, 2026.
- Tools
- No tools found
- Secrets
- None asked for
- Install scripts
- None found
- Runs as
- Node.js package
Network hosts it talks to 5
cloud.google.com· found in codecodeload.github.com· found in install commanddocs.expo.dev· found in codegithub.com· found in codelearn.duotify.com· found in code
Files it touches
~/.claude/skills/gemini-image-describer
What's inside
- Skill name
gemini-image-describer- Description
- Use this skill when a user provides an image (local file or URL) and wants a detailed Traditional Chinese description of the image content. Prefer this skill for requests like "describe this image", "what's in this picture", "OCR this screenshot", "辨識這張圖", "描述這張照片", or when extracting text from screenshots or photos.
- License
- MIT
- Instructions
- 100 lines in SKILL.md · view on GitHub
Before installing
- Ships executable scripts
Files (7)
- .gitignore
- README.md
- SKILL.md
- githubclaw.json
- package.json
- describe.js · scripts/
- describe.js · src/
SKILL.md (first lines)
# Image Describer Skill
此技能使用 Gemini Interactions API 的原生圖片理解能力,分析圖片並產生繁體中文(zh-TW)Markdown 描述,包含圖片場景描述、文字辨識(OCR)與關鍵物件列表。支援本地檔案路徑、遠端 URL 與 data URI。
## 需求條件
- `GEMINI_API_KEY` 環境變數必須已設定
- 有效的遠端圖片 URL、data URI 或可讀取的本地圖片檔案路徑
- Node.js >= 20.0.0
## 使用方式
直接執行預建置腳本 — **不需要 `npm install` 或其他額外設定**:
```sh
node .agents/skills/gemini-image-describer/scripts/describe.js <image-path-or-url>
```
### 範例
```sh
GEMINI_API_KEY=your_api_key node .agents/skills/gemini-image-describer/scripts/describe.js "https://example.com/photo.jpg"
```
```sh
GEMINI_API_KEY=your_api_key node .agents/skills/gemini-image-describer/scripts/describe.js "./photos/screenshot.png"
```
```sh
GEMINI_API_KEY=your_api_key node .agents/skills/gemini-image-describer/scripts/describe.js "data:image/png;base64,iVBOR..."
```
## Dry Run
設定 `IMAGE_DESCRIBER_DRY_RUN=1` 可在不呼叫 Gemini API 的情況下,預覽解析後的輸入 metadata:
```sh
IMAGE_DESCRIBER_DRY_RUN=1 node .agents/skills/gemini-image-describer/scripts/describe.js "https://example.com/photo.jpg"
```
## 輸出格式
輸出為繁體中文 Markdown,包含以下區段:
- **圖片描述** — 整體場景與內容描述
- **文字辨識(OCR)** — 圖片中可辨識的文字
- **關鍵物件** — 畫面中的關鍵物件列表
## Instructions for the Agent
⚠️ skill 腳本位於 **repo 根目錄**。若 cwd 不在 repo root,先獨立執行 `git rev-parse --show-toplevel` 取得路徑,再 `cd` 到該路徑後執行。禁止使用 `$(...)` 語法。
⚠️ **必須執行腳本**,不可用自身視覺能力直接回答圖片內容。所有圖片分析結果必須來自腳本輸出。
1. 向使用者取得圖片 URL 或本地檔案路徑(若尚未提供)。
2. 確認環境中已設定 `GEMINI_API_KEY`。
3. **必須**執行描述腳本,並將輸出直接寫入 `artifacts/{issue-comment-id}/result.md`:
```sh
mkdir -p artifacts/{issue-comment-id}
node .agents/skills/gemini-image-describer/scripts/describe.js "<image-path-or-url>" > artifacts/{issue-comment-id}/result.md
```
4. 若輸入為本地檔案路徑或 `file://` URL,腳本會自動轉換為 Base64 data URI。
5. 讀取 `artifacts/{issue-comment-id}/result.md` 並將內容呈現給使用者。
6. 若腳本以非零代碼結束,回報 stderr 錯誤訊息,不可自行補充圖片描述。
## 限制
- 支援的圖片格式:JPEG、PNG、GIF、WebP、BMP、SVG
- 本地大型圖片檔案會轉為 Base64,可能消耗較多記憶體
- 遠端 URL 必須可被 Gemini API 直接存取
- OCR 準確度取決於圖片品質與文字清晰度
## 錯誤處理
- 若 `GEMINI_API_KEY` 未設定,腳本會以代碼 1 結束並印出錯誤訊息。
- 若圖片輸入缺失或無效,腳本會以代碼 1 結束並印出用法說明。
- 若本地檔案無法讀取,腳本會以代碼 1 結束並回報路徑相關錯誤。
- API 錯誤會輸出到 stderr,程序以代碼 1 結束。
## 實作備註
- `scripts/describe.js` 是預建置的零依賴 bundle(由 Bun 從 `src/describe.js` 建置)。
- 模型:`gemini-3-flash-preview`(快速、平衡,適合圖片理解)。
- 遠端 URL 直接作為 `image` part 的 `uri` 傳入。
- 本地檔案轉換為 Base64 `data:<mime>;base64,...` URI 後傳送。
- 設定 `IMAGE_DESCRIBER_DRY_RUN=1` 可預覽解析後的輸入 metadata,不呼叫 Gemini API。
- 輸出以串流方式寫入 stdout,即時顯示。
- 需要 Node.js >= 20.0.0。
## 從原始碼重新建置
若需修改腳本,編輯 `src/describe.js` 後重新建置:
```sh
cd .agents/skills/gemini-image-describer
bun install
bun build src/describe.js --outfile scripts/describe.js --target node --minify
```From duotify/GitHubClawToolkit, MIT.
More from duotify/GitHubClawToolkit
- 0.754score
gemini-deep-researcher
duotify
Use this skill when a user wants a comprehensive research report on any topic with cited sources.
SkillBFreshness: B★ 254Claude Code +9 - 0.754score
gemini-summary
duotify
Use this skill when a user provides a web page URL, PDF file, video, or audio file and wants a Traditional Chinese summary.
SkillBFreshness: B★ 254Claude Code +9 - 0.754score
gemini-audio-transcriber
duotify
Use this skill when a user provides an audio file (local file or URL) and wants it transcribed into Traditional Chinese text.
SkillBFreshness: B★ 254Claude Code +9 - 0.690score
pr-creator
duotify
Use this skill when asked to create a pull request (PR).
SkillCFreshness: C★ 254Claude Code +9 - 0.621score
gemini-nanobanana
duotify
Use this skill when users ask to generate, edit, or compose images with Gemini Nano Banana 2, including text-to-image, image editing, multi-image composition…
SkillDFreshness: D★ 254Claude Code +9 - 0.621score
telegram-notify
duotify
Send Telegram Bot notifications for development events, CI/CD pipeline status, and AGENTS.md progress updates.
SkillDFreshness: D★ 254Claude Code +9 - 0.621score
google-stitch
duotify
Use this skill when users want to generate UI mockups, landing pages, wireframes, design concepts, marketing sections, or page prototypes from a prompt and…
SkillDFreshness: D★ 254Claude Code +9 - 0.621score
felo-x-search
duotify
Search X (Twitter) data using Felo X Search API.
SkillDFreshness: D★ 254Claude Code +9 - 0.621score
felo-search
duotify
Felo AI real-time web search for questions requiring current/live information.
SkillDFreshness: D★ 254Claude Code +9 - 0.621score
felo-slides
duotify
Generate PPT/slides with Felo PPT Task API in Claude Code.
SkillDFreshness: D★ 254Claude Code +9 - 0.618score
githubclaw-json-generator
duotify
為 GitHubClawToolkit 的 skill 或 template 目錄生成 githubclaw.json metadata 檔案。當使用者新增、更新 skill/template 並需要建立或更新 githubclaw.json 時觸發。也適用於批次為多個 skill/template 產生…
SkillDFreshness: D★ 254Claude Code +9 - 0.618score
felo-web-fetch
duotify
Extract webpage content with Felo Web Extract API.
SkillDFreshness: D★ 254Claude Code +9
Signals
- Stars
- 254
- Forks
- 54
- Open issues
- 1
- Last commit (this folder)
- Mar 31, 2026 (7 months ago)
From GitHub (fetched Oct 24, 2026).
Details
- License
- MIT
- Repository
- github.com/duotify/GitHubClawToolkit
- First seen
- Oct 24, 2026
- Last checked
- Oct 24, 2026
- Source
- GitHubgithub-graphql
More from duotify
- 0.754score
gemini-deep-researcher
duotify
Use this skill when a user wants a comprehensive research report on any topic with cited sources.
SkillBFreshness: B★ 254Claude Code +9 - 0.754score
gemini-audio-transcriber
duotify
Use this skill when a user provides an audio file (local file or URL) and wants it transcribed into Traditional Chinese text.
SkillBFreshness: B★ 254Claude Code +9 - 0.754score
gemini-summary
duotify
Use this skill when a user provides a web page URL, PDF file, video, or audio file and wants a Traditional Chinese summary.
SkillBFreshness: B★ 254Claude Code +9 - 0.690score
pr-creator
duotify
Use this skill when asked to create a pull request (PR).
SkillCFreshness: C★ 254Claude Code +9 - 0.621score
felo-search
duotify
Felo AI real-time web search for questions requiring current/live information.
SkillDFreshness: D★ 254Claude Code +9 - 0.621score
felo-slides
duotify
Generate PPT/slides with Felo PPT Task API in Claude Code.
SkillDFreshness: D★ 254Claude Code +9
Cite this page
gemini-image-describer skill. Agent Mart. https://agentmart.dev/skills/duotify-gemini-image-describer (last updated 2026-10-24)