Token导航 LogoToken导航

Agent Skill · 开发

scratch-blocks(暂存块)

scratch-blocks 用于补充开发相关能力,适合在 OpenClaw 中需要让 Agent 承接开发相关任务时使用。可结合来源仓库、安装命令和原始 README 继续核验具体用法。安装前建议确认权限范围、维护状态,以及是否会触发联网、命令执行或文件读写。

scratch-blocks 是开发类 Agent Skill,可用宿主为 OpenClaw,通过ClawHub获取,权限为执行命令;许可证为 MIT-0,第三方扫描已通过,按安装量在开发类的 21,746 个 Skill 里排第 9,562。

总安装 2,644周安装 108GitHub Stars 1下载 847核验 2026-05-01

安装说明

本站只整理中文说明,不托管安装包,也不代用户安装。
GitHub
许可证
MIT-0
版本
0.0.2

安装方式

通过对话安装

复制提示词发给支持本地命令或 Skills 的 AI 助手,先确认命令和权限,再让它执行。

请帮我安装这个 Agent Skill:scratch-blocks(暂存块)
来源仓库:https://github.com/z-bra0/scratch-blocks
安装命令:
openclaw skills install scratch-blocks
安装前请先检查当前环境是否支持对应 CLI,并向我确认将要执行的命令、安装目录、联网范围和文件读写权限;确认后再执行。

命令行安装

复制命令到本机终端执行。该命令会通过 OpenClaw 从第三方来源获取 Skill;本站只展示命令,不托管安装包,也不自动执行。

ClawHubOpenClaw
openclaw skills install scratch-blocks

这个 Skill 是做什么的

开发类 Skill 面向写代码、改代码和排查问题,一般会读写仓库文件,部分还会执行构建或测试命令。

当用户上传“.sb3”或“.sprite3”文件时,或者当对话是关于 Scratch 时,使用此技能,更清晰的块式可视化会有所帮助。

上手步骤

4 步
  1. 1

    确认自己的宿主工具

    公开资料标注支持 OpenClaw;先确认自己用的是其中之一,再按对应工具的 Skills 目录放置。

  2. 2

    拿到安装命令

    可以直接复制的命令是 openclaw skills install scratch-blocks;命令里的占位符换成自己的路径或密钥再执行。

  3. 3

    安装后重启宿主

    命令执行完要完全退出宿主工具再打开,让新装的 Skill 被重新扫描到,否则列表里可能看不到。

  4. 4

    按最小权限试用

    它需要执行命令,先在测试目录或测试账号里跑,确认没问题再放到正式项目。

常见问题

6 条

scratch-blocks 怎么安装?

在终端执行 openclaw skills install scratch-blocks。装完完全退出宿主工具再打开,让它重新扫描本地 Skill。

scratch-blocks 支持哪些 AI 工具?

公开资料标注支持 OpenClaw。其他工具只要能读取本地 Skill 目录,一般也能手动放进去,但要自己确认触发方式是否一致。

scratch-blocks 需要什么权限?

权限级别是执行命令。安装流程涉及命令执行,可能通过 openclaw skills install scratch-blocks 联网下载 Skill 或依赖。用户安装前应确认命令来源、仓库内容和执行环境。

scratch-blocks 是免费的吗?

来源标注的许可证是 MIT-0。本站不代收任何费用,也不托管安装包;能否商用、要不要署名,以许可证原文为准。

scratch-blocks 用的人多吗?

公开数据是累计安装 2,644 次,近一周安装 108 次,下载 847 次,GitHub 1 星。这些数字来自第三方目录统计,只能当作热度参考,不代表质量或安全性。

scratch-blocks 安全吗?

第三方扫描结果为通过,但这只覆盖已知的静态检查项,不能替代自己读一遍源码和安装命令。

值得选的地方和要先想清楚的地方

值得选的地方

  • 专注 OpenClaw 一种宿主,触发方式和使用习惯不用来回切换。
  • 许可证明确为 MIT-0,商用和分发边界好判断。
  • 第三方扫描已通过,常见的高风险行为没有命中。

先想清楚的地方

  • 需要执行命令,装到常用环境前要先评估影响面。
  • 第三方目录整理,版本更新和行为细节以来源仓库为准。

适不适合你

更适合

  • 已经在用 OpenClaw,想直接扩展它能力的人。
  • 需要处理开发类任务、又不想每次都手写提示词的人。
  • 需要许可证清楚、并且有第三方扫描结果的团队项目。

不太适合

  • 不能接受 Skill 在你机器上执行命令的人。
  • 需要厂商级技术支持和 SLA 的生产系统。

开发类 Skill 对比

6 个同类
Skill总安装GitHub 星权限许可证扫描
skill-vetter

技能审查员 · OpenClaw

8,779,018986操作浏览器MIT-0通过
ontology

本体论 · OpenClaw

6,636,960576只读MIT-0通过
entra-app-registration

内部应用程序注册 · Codex、Claude

6,631,872764接触敏感数据未明确通过
entra-app-registration

内部应用程序注册 · Codex、Claude

3,757,680201接触敏感数据未明确通过
nano-banana-pro

纳米香蕉专业版 · OpenClaw

3,526,898364接触敏感数据MIT-0有提醒
scratch-blocks

暂存块 · 本页

2,6441执行命令MIT-0通过

按安装量取同分类前几名;数字来自第三方目录统计,只作热度参考,不代表质量结论。

SKILL.md 原文

name
scratch-blocks
description
>
metadata
author
Z-Bra
version
0.0.2

Scratch Code Assistant

Output Contract

  • scratch-json is internal-only. Never paste it into the final user reply.
  • If you show Scratch code or block structure to the user, you must run scripts/render_ascii.py and use its exact output.
  • Use scratch-json only for tool input and internal reasoning.
  • If you mention Scratch code, show rendered blocks instead of describing the code only in prose whenever practical.
  • Use rendered block output in fenced code blocks.
  • Do not add a follow-up like "Want me to render an example?".
  • Do not rewrite, add, split, or restyle the rendered Scratch blocks.
  • Do not hand-draw, imitate, or approximate Scratch block ASCII from memory.

Goal

Help with Scratch code questions by representing Scratch projects into scratch-json when needed, reasoning over that internal representation, and showing Scratch code to the user through scripts/render_ascii.py instead of raw JSON. For conceptual questions, prefer a visual rendered example over prose-only explanation.

Repo Usage

  • Read files under references/ when you need skill guidance or file-handling instructions.
  • Files under data/ are runtime assets used by scripts/, not AI-facing reference docs.
  • Do not read files under data/ as reference material.

Internal scratch-json Reference

scratch-json is an internal working format for the AI. Do not return it to the user.

scratch-json is a flat JSON array of objects. Each object repeats its owning target, which can represent a Sprite or the Stage.

This is different from raw Scratch project.json / sprite.json. If you have a raw Scratch file, run scripts/extract.py first.

Each object includes:

FieldTypeDescription
typestringscript, variable, or list
targetstringOwning sprite or stage name

variable Structure

FieldTypeDescription
namestringVariable name
valueanyCurrent value

list Structure

FieldTypeDescription
namestringList name
itemslistCurrent list items

script Structure

FieldTypeDescription
blockslist of block objectsOne linked top-level script

Block Structure

Every block must include opcode. params and blocks are optional.

FieldTypeDescription
opcodestringBlock type, Scratch block opcode
paramslistOptional positional input values
blockslist of scriptsOptional sub-script branches

params Item Types

TypeExample
number / string10, "Hello!", "space"
reporter block{ opcode: motion_xposition }
variable / list / broadcast{ type: variable, name: score }

blocks Branches

  • Most blocks omit blocks.
  • control_repeat and control_if use one branch for the body.
  • control_if_else uses two branches: index 0 for then, index 1 for else.

Format

This format is for internal reasoning and tool input, not for user-facing replies.

Internal object kinds:

script, variable, list

Display

  • Must use scripts/render_ascii.py before showing Scratch code to the user
  • Treat scratch-json as intermediate data only
  • If the output is for the user, render first and reply with the rendered blocks, not JSON
  • Wrap rendered block output in fenced code blocks so it is visually separated from the explanation
# Preferred: write scratch-json to a temp file, then render that file.
# This avoids shell quoting issues and very long CLI arguments.
tmp_json=/tmp/scratchcode/blocks.json
python3 <SKILL_DIR>/scripts/render_ascii.py "$tmp_json"

# For scratch-json file path, optionally narrowed to target names:
python3 <SKILL_DIR>/scripts/render_ascii.py "<SCRATCH_JSON_PATH>" --targets Sprite1 Stage

Workflow

Step 1. Decide whether file extraction is needed

  • If the user is asking a general Scratch code question, go to step 2 and prefer a rendered example plus brief explanation.
  • If the user provides a .sb3 or .sprite3 file, follow the process in

references/UPLOADS.md.

  • If you are given raw Scratch project.json or sprite.json, run scripts/extract.py first before scripts/render_ascii.py.

Step 2. Reason Internally

  • Use scratch-json only as an internal representation for analysis.
  • For simple conceptual questions, create a minimal internal example when needed so you can render it.
  • When you create scratch-json yourself, write it to a temp file and pass the file path to scripts/render_ascii.py.

Step 3. Reply to the User

  • Never return scratch-json data to the user.
  • For "how to" or "what does this do" questions, prefer rendered Scratch blocks first.
  • If you want to show Scratch code, always run scripts/render_ascii.py first.
  • Put rendered Scratch output inside a fenced code block.
  • If you did not run render_ascii.py, do not output boxed ASCII at all; answer in prose or run the renderer first.
  • After rendering, explain the answer in short prose around the rendered block output when useful.
  • Before sending the final answer, check: "Am I about to paste scratch-json?" If yes, stop and render it first.

平台分布

OpenClaw

79.83%
按下载量换算676

安全审计

VirusTotal

通过

ClawScan

通过

Static analysis

通过

权限和风险

执行命令

安装流程涉及命令执行,可能通过 openclaw skills install scratch-blocks 联网下载 Skill 或依赖。用户安装前应确认命令来源、仓库内容和执行环境。

安装前确认

本站仅展示第三方公开信息,不托管安装包,不提供自动安装或运行环境。安装前应自行审查源码、依赖和命令行为。目前只有一个公开来源,建议同时核对源仓库。

继续浏览同类 Skills