← 返回 Skills 市场
wondershare-boop

Studio Ghibli Image Filter

作者 wondershare-boop · GitHub ↗ · v1.0.3 · MIT-0
cross-platform ✓ 安全检测通过
186
总下载
0
收藏
0
当前安装
4
版本数
在 OpenClaw 中安装
/install mediaio-ghibli-image-filter
功能描述
Transform photos and images into Studio Ghibli-style artwork using AI via Media.io OpenAPI. Applies the iconic hand-drawn, painterly aesthetic of Ghibli film...
使用说明 (SKILL.md)

\r \r

Media.io Ghibli Image Filter Skill\r

\r

Overview\r

This skill calls Media.io OpenAPI to run Ghibli-style conversion using model code effects-ghibli.\r The API is asynchronous:\r

  1. Submit generation request and get task_id.\r
  2. Poll task result endpoint until the task is finished.\r \r

When To Use\r

  • The user wants to convert an image to Ghibli-like style.\r
  • The user can provide an image URL reachable by Media.io servers.\r
  • The user wants task-based generation with polling.\r \r

When Not To Use\r

  • The user asks for local file upload only (this API expects image URL input).\r
  • The user asks for non-Media.io providers.\r
  • The user asks for real-time synchronous image output in one call.\r \r

Requirements\r

\r

Environment Variables\r

| Variable | Required | Description |\r |---|---|---|\r | MEDIAIO_API_KEY | Yes | Media.io OpenAPI key, used in header X-API-KEY. |\r \r

Base Headers\r

  • Content-Type: application/json\r
  • X-API-KEY: $MEDIAIO_API_KEY\r \r

Supported Endpoints\r

\r

1) Query Credits\r

  • Method: POST\r
  • Endpoint: https://openapi.media.io/user/credits\r
  • Body: {}\r
  • Purpose: check available credits before generation.\r \r

2) Create Ghibli Task\r

  • Method: POST\r
  • Endpoint: https://openapi.media.io/generation/effects/effects-ghibli\r
  • Body:\r
{\r
	"data": {\r
		"images": "https://example.com/input.jpg",\r
		"ratio": "9:16",\r
		"batch": "1"\r
	}\r
}\r
```\r
- Required fields:\r
	- `data.images` (string URL)\r
- Optional fields:\r
	- `data.ratio` (string): `9:16`, `16:9`, `1:1`, `4:3`, `3:4`, `3:2`, `2:3`\r
	- `data.batch` (string): `1`, `2`, `3`, `4`\r
\r
### 3) Query Task Result\r
- Method: `POST`\r
- Endpoint: `https://openapi.media.io/generation/result/{task_id}`\r
- Body: `{}`\r
- Path parameter:\r
	- `task_id` (string, required)\r
\r
## Request and Response Contract\r
\r
### Common Success Envelope\r
```json\r
{\r
	"code": 0,\r
	"msg": "",\r
	"data": {},\r
	"trace_id": "..."\r
}\r
```\r
\r
### Create Task Response\r
- On success, `data.task_id` is returned.\r
\r
### Task Result Response\r
- `data.status` can be one of the following values:\r
	- `waiting`: queued\r
	- `processing`: running\r
	- `completed`: completed successfully\r
	- `failed`: failed\r
	- `timeout`: timed out\r
- `data.reason`: provides additional context (e.g., `success` or error message)\r
- When status is `completed`:\r
	- `data.result` is an array of output objects with generated URLs\r
	- Each result object contains `val` (internal path), `preview` (public HTTPS URL), and `status` (completion status)\r
\r
## Standard Invocation Flow\r
1. Call `user/credits` to verify balance.\r
2. Call `effects-ghibli` with `data.images` and optional `data.ratio`, `data.batch`.\r
3. Extract `task_id`.\r
4. Poll `generation/result/{task_id}` every 3 to 5 seconds.\r
5. Stop when status is `completed` or `failed`.\r
6. Return output URLs from `data.result` when `completed`.\r
\r
## cURL Examples\r
\r
### Query Credits\r
```bash\r
curl --request POST \\r
	--url https://openapi.media.io/user/credits \\r
	--header 'Content-Type: application/json' \\r
	--header "X-API-KEY: $MEDIAIO_API_KEY" \\r
	--data '{}'\r
```\r
\r
### Create Ghibli Task\r
```bash\r
curl --request POST \\r
	--url https://openapi.media.io/generation/effects/effects-ghibli \\r
	--header 'Content-Type: application/json' \\r
	--header "X-API-KEY: $MEDIAIO_API_KEY" \\r
	--data '{\r
	"data": {\r
		"images": "https://example.com/input.jpg",\r
		"ratio": "9:16",\r
		"batch": "1"\r
	}\r
}'\r
```\r
\r
### Query Task Result\r
```bash\r
curl --request POST \\r
	--url https://openapi.media.io/generation/result/\x3Ctask_id> \\r
	--header 'Content-Type: application/json' \\r
	--header "X-API-KEY: $MEDIAIO_API_KEY" \\r
	--data '{}'\r
```\r
\r
#### Successful Response Example\r
```json\r
{\r
	"code": 0,\r
	"msg": "",\r
	"data": {\r
		"task_id": "effect-86f0f82a-36dc-4a7c-928a-721a18ef482f",\r
		"status": "completed",\r
		"reason": "success",\r
		"result": [\r
			{\r
				"val": "aicloudtmp/550160908/3/202603/1/combo_tm_alg-20260317165022-802800-60eb3-dwt.png",\r
				"preview": "https://url_to_generated_image.png",\r
				"status": "completed"\r
			}\r
		]\r
	},\r
	"trace_id": "a18315ba568b5c34407808d12cbc8457"\r
}\r
```\r
\r
Response fields when status is `completed`:\r
- `data.task_id`: unique task identifier\r
- `data.status`: `completed` indicates successful completion\r
- `data.reason`: `success` indicates no error occurred\r
- `data.result`: array of output objects, each containing:\r
	- `val`: internal file path of the generated asset\r
	- `preview`: publicly accessible HTTPS URL for the generated asset\r
	- `status`: `completed` for each result item\r
\r
## Error Handling Guidance\r
- Treat `code != 0` as failure.\r
- Typical authentication errors:\r
	- `374004`: not authenticated. Apply for an APP KEY at https://developer.media.io/.\r
- Typical request validation error:\r
	- `490000`: params error\r
- Typical billing/credits error:\r
	- `490505`: insufficient credits. Recharge before invoking generation APIs.\r
- Always include `trace_id` in logs for troubleshooting.\r
\r
## Agent Behavior Requirements\r
- Validate that input contains a non-empty image URL before calling the create endpoint.\r
- Do not claim immediate output after task creation; always poll by `task_id`.\r
- If credits are insufficient, return a clear message and stop instead of retry loops.\r
- Avoid exposing raw API keys in logs or responses.\r
\r
## Safety and Compliance Notes\r
- Only process user-provided or user-authorized images.\r
- Do not imply identity verification or biometric certainty from generated images.\r
- Generated output is synthetic media and should be presented as edited content.\r
\r
## References\r
- Media.io platform: https://developer.media.io/\r
- API documentation: https://platform.media.io/docs/\r
安全使用建议
This skill appears to do exactly what it says: call Media.io to convert a publicly reachable image URL into a Ghibli-like image. Before installing, ensure you: (1) trust Media.io and the skill publisher, (2) store MEDIAIO_API_KEY with least privilege and billing limits (it grants API access and may incur charges), (3) do not send sensitive or private images accessible only locally (the API expects public URLs), (4) confirm you are comfortable with any legal/copyright implications of generating images “in the style of” Studio Ghibli, and (5) verify agents' logging or transcripts won’t leak the API key (the SKILL.md warns to avoid exposing keys, but your agent or logs may still accidentally include them). If you need to allow local files, confirm a secure proxy or upload workflow rather than giving the skill broader filesystem/network access.
功能分析
Type: OpenClaw Skill Name: mediaio-ghibli-image-filter Version: 1.0.3 The skill bundle provides legitimate instructions for an AI agent to interact with the Media.io OpenAPI to apply image filters. It follows a standard asynchronous API flow (credit check, task creation, and polling) using curl and the MEDIAIO_API_KEY environment variable, with no evidence of malicious intent, data exfiltration, or unauthorized execution.
能力评估
Purpose & Capability
Name/description match the runtime instructions: the SKILL.md documents calling Media.io endpoints (credits, create effect task, poll results) and requires MEDIAIO_API_KEY and curl — all appropriate for a cloud image-effect integration.
Instruction Scope
Instructions are scoped to submitting an image URL, polling task status, and returning output URLs. The guide explicitly forbids local-only uploads, warns about not exposing API keys in logs, and asks the agent to validate image URLs and credits. The SKILL.md does not instruct reading unrelated files or environment variables.
Install Mechanism
No install spec or code is present (instruction-only). This minimizes on-disk risk; required runtime binary is only curl, which is reasonable for the provided cURL examples.
Credentials
Only MEDIAIO_API_KEY is required and declared as the primary credential. That is proportional and necessary for authenticating to Media.io. No unrelated secrets or system paths are requested.
Persistence & Privilege
The skill does not request always:true or persistent system-wide changes. It is user-invocable and allows autonomous invocation (platform default) but does not request elevated persistence or cross-skill config access.
如何使用
  1. 确保已安装 OpenClaw(本地或 Docker 部署)
  2. 在对话框中输入安装命令:/install mediaio-ghibli-image-filter
  3. 安装完成后,直接呼叫该 Skill 的名称或使用 /mediaio-ghibli-image-filter 触发
  4. 根据 Skill 的参数说明提供必要输入,即可获得结构化输出
版本历史
v1.0.3
- Updated metadata to include OpenClaw format and declare a bins requirement for curl. - Removed legacy metadata on Media.io provider. - No changes to API behavior or invocation flow. - Documentation and usage details remain the same.
v1.0.2
- No changes detected in this release; SKILL.md content remains unchanged. - Maintains previous functionality for transforming images into Studio Ghibli-style artwork using Media.io OpenAPI.
v1.0.1
No changes detected in this version. - No file or documentation changes between versions 1.0.0 and 1.0.1.
v1.0.0
- Initial release of mediaio-ghibli-image-filter. - Convert photos into Studio Ghibli-style artwork using Media.io OpenAPI. - Supports asynchronous task-based processing and polling for completion. - Requires an image URL as input (local uploads not supported). - Checks user credits before generation; provides clear error messages on insufficient credits. - Returns public URLs for generated Ghibli-style images when processing is complete.
元数据
Slug mediaio-ghibli-image-filter
版本 1.0.3
许可证 MIT-0
累计安装 0
当前安装数 0
历史版本数 4
常见问题

Studio Ghibli Image Filter 是什么?

Transform photos and images into Studio Ghibli-style artwork using AI via Media.io OpenAPI. Applies the iconic hand-drawn, painterly aesthetic of Ghibli film... 它是一个面向 Claude Code / OpenClaw 的 AI Agent Skill 插件,目前累计下载 186 次。

如何安装 Studio Ghibli Image Filter?

在 OpenClaw 或 Claude Code 对话框中运行命令「/install mediaio-ghibli-image-filter」即可一键安装,无需额外配置。

Studio Ghibli Image Filter 是免费的吗?

是的,Studio Ghibli Image Filter 完全免费,采用 MIT-0 许可证,可自由下载、安装和使用。

Studio Ghibli Image Filter 支持哪些平台?

Studio Ghibli Image Filter 跨平台运行,可在任意部署了 OpenClaw / Claude Code 的环境中使用(cross-platform)。

谁开发了 Studio Ghibli Image Filter?

由 wondershare-boop(@wondershare-boop)开发并维护,当前版本 v1.0.3。

💬 留言讨论