← 返回 Skills 市场
156
总下载
2
收藏
0
当前安装
6
版本数
在 OpenClaw 中安装
/install word-jumble
功能描述
Generate a Word Jumble puzzle — scrambled words with circled letters that spell out a final idiom, plus a cartoon illustration hint and a printable puzzle im...
安全使用建议
This skill appears to do exactly what it says: generate, validate, and render Word Jumble puzzles. Before installing/using, consider the following: (1) The final answer is intentionally embedded in the HTML DOM as plaintext (visually rotated/small) — anyone or any tool that can access the served page (or the generated HTML file) can read the answer. If you want the answer to be secret, modify the template to remove or encode the answer server-side. (2) Image generation sends prompts to whatever external image provider the platform is configured with; do not put sensitive information into image prompts. (3) The included comment in the HTML claims the answer is ROT13-encoded, but the implementation writes the plaintext solution — this is an informational mismatch (not a security exploit). (4) The renderer starts a local HTTP server bound to 127.0.0.1 and writes temporary files; confirm you trust the agent runtime to handle these files. If you need the skill to be more private, request removal of the answer from the DOM and/or local-only rendering without writing persistent outputs.
能力评估
Purpose & Capability
Name/description (Word Jumble generator + printable output + cartoon hint) align with the included assets and scripts: a template HTML, a validator, and a renderer that prepares a local HTTP page for screenshotting. No unrelated credentials, binaries, or installs are requested.
Instruction Scope
Instructions stay within the declared purpose: generate puzzle JSON, validate with scripts/validate_puzzle.py, produce an image via the platform image_generate tool, and render via a local HTTP server for screenshot. Two caveats: (1) the HTML template embeds the final solution into the DOM in plaintext (upside-down/small), which the SKILL.md calls out — this means the answer is present in the served page and could be read programmatically even if visually obfuscated; (2) image generation uses the platform image_generate tool which sends prompts to an external image generation API (and thus prompt content is exposed to that provider). Both behaviors are intentional per the skill, but the user should be aware.
Install Mechanism
No install spec is provided (instruction-only with included scripts/assets). This is low risk: nothing is downloaded or written to system locations beyond the workspace/temp files the scripts create.
Credentials
The skill declares no required environment variables, no credentials, and no config paths. The only external interaction is the platform-managed image provider (handled by the platform, not requested by the skill). This is proportionate for an image-using puzzle generator.
Persistence & Privilege
always is false and the skill does not request persistent or elevated platform privileges, nor does it modify other skills or system-wide configs. The renderer uses a short-lived local HTTP server bound to 127.0.0.1 and temporary directories, which are cleaned up.
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 word-jumble - 安装完成后，直接呼叫该 Skill 的名称或使用
/word-jumble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
v1.2.3
- Updated the printable puzzle screenshot instructions to use a 900×900px viewport instead of 1200×900px.
- Clarified that the printable PNG is now 900×900 pixels.
- Revised printable answer key handling: the answer key is now presented upside-down and in small print at the bottom of the page, not ROT13-encoded in the DOM.
- Improved documentation wording for puzzle rendering and answer display.
- No functional or behavioral changes beyond documentation updates.
v1.2.1
- Changed printable puzzle screenshot size from 1200×900px to 900×900px, and updated instructions to use a narrower responsive layout for tighter printables.
- Documentation updated throughout to reflect the new printable dimensions.
- No other functional changes.
v1.1.0
No user-visible changes in this version.
- No file changes detected compared to the previous version.
- Documentation, usage, and functionality remain the same.
v1.0.2
- Updated documentation to clarify that the local HTTP server for rendering the printable binds to 127.0.0.1 only for increased security.
- Improved instructions for starting the server by specifying the use of the --bind 127.0.0.1 option.
- No changes to underlying functionality or puzzle format—documentation update only.
v1.0.1
- Added clarifications on image generation: the cartoon uses the platform image_generate tool, which sends prompts to an external API.
- Updated description and SKILL.md to highlight use of external providers for cartoon creation.
- Added a new Security section specifying that images are generated via API, answer keys are ROT13-encoded, and the local HTTP server is restricted to localhost.
- Other minor clarifications on expected output and generation process.
v1.0.0
- Initial release of word-jumble skill.
- Generates Word Jumble puzzles: 4 scrambled words, circled letters reveal a final idiom.
- Produces puzzle JSON, cartoon hint image, and a printable 1200×900 puzzle image.
- Cartoon is a black-and-white, hand-drawn newspaper style visual hint—no labels or words.
- Includes validation instructions and detailed output locations.
- Skill does not post outputs; caller is responsible for further actions.
元数据
常见问题
Word Jumble 是什么？
Generate a Word Jumble puzzle — scrambled words with circled letters that spell out a final idiom, plus a cartoon illustration hint and a printable puzzle im... 它是一个面向 Claude Code / OpenClaw 的 AI Agent Skill 插件，目前累计下载 156 次。
如何安装 Word Jumble？
在 OpenClaw 或 Claude Code 对话框中运行命令「/install word-jumble」即可一键安装，无需额外配置。
Word Jumble 是免费的吗？
是的，Word Jumble 完全免费，采用 MIT-0 许可证，可自由下载、安装和使用。
Word Jumble 支持哪些平台？
Word Jumble 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 Word Jumble？
由 Marv（@robin-marv）开发并维护，当前版本 v1.2.3。
推荐 Skills