image-element-remove
/image-element-remove
从已有图片里去掉水印、文字或局部物体,覆盖通用水印消除、文档半透明水印消除、局部物体消除和指定文字消除。仅在用户明确要求对当前图片做这些删除操作时触发;模糊的“去掉字”或未说明对象的泛化修图请求不单独触发。
- Author
- meitu ↗
- Repository
- meitu-skills ↗
- Licence
- MIT
- Last repository change
- 28.08.2026
Assessment and security
- Files in this package
- 1
- Scripts
- 0
- HTML comments
- 0
- URLs in SKILL.md
- 0
No signals detected. Repository changed on 28.08.2026, without references to retired models or APIs.
MIT, open: use and modify while retaining author attribution.
What it does and when to use it
从已有图片里去掉水印、文字或局部物体,覆盖通用水印消除、文档半透明水印消除、局部物体消除和指定文字消除。仅在用户明确要求对当前图片做这些删除操作时触发;模糊的“去掉字”或未说明对象的泛化修图请求不单独触发。
Triggers from the description
Original author description
从已有图片里去掉水印、文字或局部物体,覆盖通用水印消除、文档半透明水印消除、局部物体消除和指定文字消除。仅在用户明确要求对当前图片做这些删除操作时触发;模糊的“去掉字”或未说明对象的泛化修图请求不单独触发。
How it works
- Overview
- API Mapping
- Dependencies
- Core Workflow
- Output
- 基线 Task ID
SKILL.md preview
---
name: image-element-remove
description: "从已有图片里去掉水印、文字或局部物体,覆盖通用水印消除、文档半透明水印消除、局部物体消除和指定文字消除。仅在用户明确要求对当前图片做这些删除操作时触发;模糊的“去掉字”或未说明对象的泛化修图请求不单独触发。"
version: "1.0.0"
metadata: {"openclaw":{"requires":{"bins":["meitu"],"env":["MEITU_OPENAPI_ACCESS_KEY","MEITU_OPENAPI_SECRET_KEY","MEITU_OPENAPI_TOOL_TASK_MODE"],"paths":{"read":["~/.meitu/credentials.json","~/.meitu/tool-registry.json","~/.openclaw/workspace/visual/","./openclaw.yaml"],"write":["~/.openclaw/workspace/visual/","./output/"]}},"primaryEnv":"MEITU_OPENAPI_ACCESS_KEY"}}
security:
credential_use: "Uses Meitu OpenAPI credentials from env or ~/.meitu/credentials.json for CLI calls; credentials must not be echoed, logged, or embedded in prompts."
remote_processing: "User-provided image URLs and generated removal prompts are sent to Meitu OpenAPI."
persistence: "Edited images are written to the resolved local output directory."
usage_notice: "Removing watermarks, logos, text, or provenance indicators may involve legal, ethical, or authenticity risks. Confirm the user has the right to edit the image in this way."
requirements:
credentials:
- name: MEITU_OPENAPI_ACCESS_KEY
source: env | ~/.meitu/credentials.json
- name: MEITU_OPENAPI_SECRET_KEY
source: env | ~/.meitu/credentials.json
env:
MEITU_OPENAPI_TOOL_TASK_MODE: command
permissions:
- type: file_read
paths:
- ~/.meitu/credentials.json
- ~/.meitu/tool-registry.json
- ~/.openclaw/workspace/visual/
- ./openclaw.yaml
- type: file_write
paths:
- ~/.openclaw/workspace/visual/
- ./output/
- type: exec
commands:
- meitu
---
# 图片消除(image-element-remove)
## Overview
对已有图片做元素消除(去水印/去文字/去局部物体),覆盖 4 类场景:通用水印消除(一键去水印含半透明)、文档场景半透明水印、局部物体/元素/指定文字消除、降级兜底链。路由铁律:用户指定具体文字内容时必须使用 `praline_pro`(路径 C),禁止走通用水印/全部文字路径(否则会清除全部文字)。
执行前应让用户清楚知道:本 Skill 会读取 Meitu 凭证、调用本地 `meitu` CLI、将用户提供的图片与生成的消除指令发送到 Meitu OpenAPI 处理,并把结果写入 `./output/` 或 `$VISUAL/output/image-element-remove/`。若请求涉及去水印、去 Logo、去认证/署名信息或类似来源标识,应提醒用户确认自己具备合法编辑权限。
## API Mapping
- 通用水印/全部文字消除:通用消除路径。
- 文档场景半透明水印:文档水印路径。
- 局部物体或指定文字消除:优先 `praline_pro`,失败依次降级为 `praline_lite`、`mint_edit`。
## Dependencies
- **meitu-cli**: `>=2.0.6`(`npm install -g meitu-cli@latest`)
- **凭证**:CONFIG AKSK → `meitu tools update`;EXEC AKSK → 实际执行(见根 `CONFIG.md`)
- **环境变量**:`MEITU_OPENAPI_TOOL_TASK_MODE=command`
> 路径别名:`$VISUAL` = `{OPENCLAW_HOME}/workspace/visual/`
## Core Workflow
```
Preflight → Execute → Deliver
```
### Preflight
1. `meitu --version` ≥ 2.0.6(否则 `npm install -g meitu-cli@latest`)
2. 确认已跑过 `meitu tools update`(用 CONFIG AKSK)
3. 当前 AKSK = EXEC,且 `MEITU_OPENAPI_TOOL_TASK_MODE=command`
4. 解析 output_dir:openclaw.yaml → `./output/` |else → `$VISUAL/output/image-element-remove/`;`mkdir -p`
### Execute
**触发信号 / 路由规则**
核心判断维度:**前置判断(指定文字?) + 消除内容类型**。
决策顺序:
1. **前置判断(路由铁律)**:用户引用/描述了特定文字内容?
- ✅ 是 → **强制使用** `praline_pro`(禁止走通用消除路径)
- ❌ 否 → 继续分类
2. 按消除内容分支:
- 路径 A:去水印 / 去全部文字(全量词) → 通用消除路径
- 路径 B:明确文档场景半透明水印 → 文档水印路径
- 路径 C:局部物体/指定文字 → `praline_pro` → 降级链
| 场景 | 判定关键词 | 路由 |
|------|----------|------|
| 指定具体文字内容(铁律) | 去掉'xxx'、去除文字'xxx' | **`praline_pro`** |
| 通用水印 | 去水印、去掉水印、去半透明水印 | 通用消除路径,target=watermark |
| 全部文字(含全量词) | 去全部文字、所有文字、一键去文字 | 通用消除路径,target=text |
| 文档场景半透明水印(用户原话明确) | 文档水印、扫描件水印、小说水印 | 文档水印路径 |
| 局部物体/元素 | 去路人、去 Logo、消除杂物、去角标 | `praline_pro` → 降级链 |
| 意图模糊("去掉字") | 未说全部也未指定具体内容 | 轻确认 |
**显式约束消除原则**:`praline_pro`、`praline_lite` 和 `mint_edit` 的消除 prompt 必须使用**四段式模板**(SCOPE + PRESERVE + REFLOW + NEGATIVE),显式锁定剩余元素 12 项视觉属性,防止模型顺手改动非目标元素。
**参数定义**
通用参数:
| 参数 | 类型 | 必填 | 说明 |
|------|------|------|------|
| `image_url` | STRING | 是 | 图片地址。缺失 → 提示"请提供需要处理的图片" |
通用消除路径的额外参数:
| 参数 | 类型 | 必填 | 范围 | 默认值 | 说明 |
|------|------|------|------|--------|------|
| `target` | STRING | 否 | watermark / text | watermark | 消除目标类型 |
局部消除模型的额外参数:
| 参数 | 类型 | 必填 | 说明 |
|------|------|------|------|
| `prompt` | STRING | 是 | 按四段式模板(SCOPE + PRESERVE + REFLOW + NEGATIVE)生成的英文消除指令 |
**四段式 prompt 模板**:
```
Task: Element removal.
Remove ONLY the specified element from the image: {target_description}.
STRICTLY PRESERVE all remaining visual elements, including: font family, size, color, weight, style, stroke, shadow, position, alignment, background, lighting, composition, and any other text or objects not targeted for removal.
If the removal affects part of a text segment, REFLOW the remaining text to read as a natural, complete phrase — close any gaps, maintain baseline and alignment, preserve visual balance and typographic consistency.
DO NOT modify any other elements in the image.
DO NOT change the overall color palette, lighting, composition, or style.
```
**工具调用**
```bash
# 通用水印
meitu image-element-remove --image_url <url> --target watermark --json --download-dir {output_dir} --skill_name skill_image-element-remove
```
```bash
# 指定文字/局部物体
meitu image-element-remove --image_url <url> --prompt "<four_segment_prompt>" --model praline_pro --json --download-dir {output_dir} --skill_name skill_image-element-remove
```
**错误降级**
| 场景 | 处理方式 |
|------|------|
| `image_url` 缺失 | 提示"请提供需要处理的图片" |
| `image_url` 不可访问 | 直接返回无效错误,不重试 |
| 通用消除路径失败 + 用户提到"半透明" | 降级文档水印路径 → `praline_pro` → `praline_lite` → `mint_edit` |
| 通用消除路径失败(非半透明) | 返回错误,不盲目降级 |
| 文档水印路径失败 | 降级 `praline_pro` → `praline_lite` → `mint_edit` |
| `praline_pro` 失败 | 自动降级 `praline_lite` |
| `praline_lite` 失败 | 自动降级 `mint_edit` |
| `mint_edit` 失败 | 返回错误 |
| 未检测到水印/文字 | 返回原图,告知未检测到 |
| 内容合规拦截 | 返回合规提示,不重试、不降级 |
### Deliver
- 直接使用 Preflight 解析的 output_dir
- 从 `downloaded_files[0].saved_path` 读取已下载文件路径
- `mv {downloaded_files[0].saved_path} {output_dir}/{YYYY-MM-DD}_{descriptive}_image-element-remove.jpg`
## Output
- **格式**:JPEG/PNG(保持原格式)
- **命名**:`{YYYY-MM-DD}_{descriptive}_image-element-remove.jpg`
- **位置**:项目 → `./output/`,一次性 → `$VISUAL/output/image-element-remove/`
## 基线 Task ID
见 `references/task-id-baseline.md` 中对应行。
Versions
The database has no other skills named image-element-remove.
Database snapshot: 13.11.2026 · package version e1b4dbaf6156 (SHA-256 digest of the entire package)
Downloads in the last 30 days: 0 · ranking
Report a problem with this skill: harmful content, incorrect descriptions or an author’s removal request.
Similar skills
- 100video-element-remove
使用 meitu-cli 从视频中消除路人/字幕/水印/Logo 或一键清屏全部文字。当用户提到去路人、去行人、去字幕、去对白、去水印、去 Logo、去角标、消除抖音号、清屏、纯净画面、不要出现任何字时触发。 - 100image-studio
Create images from a text description, or edit photos and graphics the user provides (remove backgrounds or watermarks, retouch, restyle, in-paint). Can produce multiple variants when the user wants options to choose from. - 100image-generation
Use this skill for any image-related AI generation or editing task. Triggers include: GENERATE: "generate image", "create image", "make picture", "draw", "visualize", "image of", "create art", "generate art" EDIT: "edit image", "modify image", "change image", "update image",… - 100li-human
Strip the machine fingerprint out of any draft - em dashes, AI slop words, invisible watermark characters - and score it against a five-check detection panel before it goes out. Use whenever text needs to sound human, when the user says humanize, "does this sound like AI",… - 100removing-subtitles
Removes hardcoded subtitles from videos using AI inpainting. Use when user wants to remove subtitles, erase text from video, clean video from captions, delete burned-in subtitles, remove video watermarks, clean hardcoded text, or strip embedded subtitles. 去字幕、去除字幕、删除字幕、清除字幕、去硬字幕、去水印、擦除字幕、移除字幕。 - 100linkpix-media-tools
结合本地工具提供视频和图片的智能处理能力:视频去水印、去字幕、超清修复(超分),图片抠图白底、换背景、元素消除、文字修改、加水印、压缩,帮助电商卖家快速完成素材优化与二次创作。当用户要求处理/优化/清理视频或图片素材时必须触发。关键词:LinkPix、qhkit、视频处理、图片处理、去水印、去字幕、超分、画质提升、抠图、换背景、消除、图片压缩、加水印、素材优化、二次创作。