← 返回 Skills 市场
Weryai Image Toolkits
作者
WeryAI-Developer
· GitHub ↗
· v0.1.6
· MIT-0
182
总下载
0
收藏
0
当前安装
7
版本数
在 OpenClaw 中安装
/install weryai-image-toolkits
功能描述
Use when the user needs WeryAI image tools to analyze and transform existing images. Generate reusable prompts, convert and optimize visuals via background r...
安全使用建议
This package appears coherent for WeryAI image editing, but review and act cautiously: 1) Provide only a WERYAI API key you trust; the tool will send your images (including any local file paths you pass) to the configured WERYAI_BASE_URL. 2) Avoid giving local paths to sensitive files; the uploader supports many file types (images, text, etc.) and will POST them to the API. 3) Do not change WERYAI_BASE_URL unless you trust the destination—overriding it can make the client send sensitive data to an arbitrary host (the code warns about this). 4) Use the built-in dry-run and inspect the scripts (scripts/image_toolkits.js and scripts/vendor/weryai-core/*) before running paid jobs, and run first paid jobs from an isolated environment or short-lived session. If you want extra assurance, test only with public HTTPS image URLs (not local files) and verify account balance and API key permissions first.
功能分析
Type: OpenClaw Skill
Name: weryai-image-toolkits
Version: 0.1.6
The skill provides image processing tools via the WeryAI API but exhibits high-risk behaviors, specifically the ability to read and upload arbitrary local files to a remote endpoint (api.weryai.com). While this is plausibly needed for processing local images, the implementation in 'scripts/image_toolkits.js' lacks directory restrictions or robust file-type validation, creating a significant path for data exfiltration if the agent is manipulated. Furthermore, 'SKILL.md' and 'AGENTS.md' explicitly instruct the AI agent to hide internal status commands from the user, and the script contains misleadingly named functions like 'validateHttpsUrl' that do not actually enforce HTTPS, increasing the overall risk profile.
能力评估
Purpose & Capability
Name/description, declared env vars (WERYAI_API_KEY, WERYAI_BASE_URL), required binary (node), and the included scripts all implement image-editing and upload/processing workflows described in SKILL.md. The requested items are proportional to the stated purpose.
Instruction Scope
Runtime instructions and the CLI code operate on image URLs and local file paths, upload local files to the WeryAI service, support dry-run, and poll task status. This is appropriate for an image-editing toolkit. Caveat: the code permits overriding WERYAI_BASE_URL (and will upload to whatever host that points to) and will upload arbitrary allowed local files you provide; if misconfigured (or if you accidentally supply a sensitive path) that could leak local content. The SKILL.md warns about trusting overrides and recommends reviewing scripts before production use.
Install Mechanism
No network install spec is included; the package contains Node scripts to run locally. There are no external download URLs or archive extracts. Requiring node is expected and proportionate.
Credentials
Only the WERYAI_API_KEY is a secret and is the declared primary credential; WERYAI_BASE_URL is expected for host override. The code also reads optional env vars (e.g., WERYAI_POLL_INTERVAL_MS, WERYAI_POLL_TIMEOUT_MS, WERYAI_ALLOW_INSECURE_UPLOAD) that are not listed as required but are benign operational knobs. The number and type of env vars requested are proportionate to the skill's function.
Persistence & Privilege
always:false and no install-time modifications to other skills or global agent settings. The skill runs as a normal, user-invocable module and does not request permanent elevated privileges.
如何使用
- 确保已安装 OpenClaw（本地或 Docker 部署）
- 在对话框中输入安装命令：
/install weryai-image-toolkits - 安装完成后，直接呼叫该 Skill 的名称或使用
/weryai-image-toolkits触发 - 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
v0.1.6
weryai-image-toolkits 0.1.6
- No code or documentation changes detected in this version.
- Functionality and usage remain unchanged from the previous release.
v0.1.5
weryai-image-toolkits 0.1.5
- Added core vendor scripts for image toolkit functionality: balance, client, errors, model-display, normalize, status, upload, and wait.
- Updated documentation to clarify direct support for local file path uploads and mention WERYAI_BASE_URL as an optional override.
- Now lists both WERYAI_API_KEY and WERYAI_BASE_URL as required environment variables.
- Updated CLI examples and improved instructions for local usage and file handling.
v0.1.4
weryai-image-toolkits v0.1.4
- Updated usage guidance to recommend explaining capabilities in natural language and giving example prompts, rather than sharing raw shell commands when users ask for help.
- Adjusted "Preferred Commands" to focus on core operations and removed full shell usage blocks from general guidance.
- Clarified when to provide operator-style guidance versus step-by-step shell instructions.
- No functional or code changes; documentation and user messaging improvements only.
v0.1.3
- Removes all vendor/weryai-core helper scripts from the codebase, simplifying dependencies.
- No user-facing features or APIs are changed.
- Documentation and primary skill functionality remain unchanged.
- Internal code is now lighter, easier to maintain, and less complex.
v0.1.2
weryai-image-toolkits v0.1.2
- Added support for local/file image sources, which are now uploaded before processing (previously only public HTTPS URLs allowed).
- Added new vendor scripts: model-display.js and upload.js, improving image handling and input options.
- Updated documentation to reflect support for both http/https and local file sources as inputs.
v0.1.1
- Major cleanup: removed 31 vendor/shared and weryai-image script and reference files, streamlining the codebase.
- Retained and updated only core toolkit scripts needed for post-processing workflows.
- No impact on user prompts, main features, or tool routing.
- Documentation remains focused and accurate, reflecting the reduced file footprint.
v0.1.0
weryai-image-toolkits 0.1.0
- Initial release for official WeryAI image editing and post-processing on existing images via public HTTPS URLs.
- Supports image-to-prompt, background change/remove, canvas expand, face swap, reframe, repair, text erase, image translate, and image upscale.
- Strict secret handling: requires WERYAI_API_KEY as the sole runtime secret, with robust environment and API guidance.
- Node.js 18+ runtime dependency; rejects local file uploads—public HTTPS URLs only.
- Includes detailed setup, authentication, security, tool routing, and parameter assistance documentation.
- Provides dry-run and validation support to avoid accidental paid requests.
元数据
常见问题
Weryai Image Toolkits 是什么？
Use when the user needs WeryAI image tools to analyze and transform existing images. Generate reusable prompts, convert and optimize visuals via background r... 它是一个面向 Claude Code / OpenClaw 的 AI Agent Skill 插件，目前累计下载 182 次。
如何安装 Weryai Image Toolkits？
在 OpenClaw 或 Claude Code 对话框中运行命令「/install weryai-image-toolkits」即可一键安装，无需额外配置。
Weryai Image Toolkits 是免费的吗？
是的，Weryai Image Toolkits 完全免费，采用 MIT-0 许可证，可自由下载、安装和使用。
Weryai Image Toolkits 支持哪些平台？
Weryai Image Toolkits 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 Weryai Image Toolkits？
由 WeryAI-Developer（@weryai-developer）开发并维护，当前版本 v0.1.6。
推荐 Skills