← 返回 Skills 市场
Visual Explainer Main
作者
austindixson
· GitHub ↗
· v1.0.0
347
总下载
0
收藏
1
当前安装
1
版本数
在 OpenClaw 中安装
/install visual-explainer-main
功能描述
Generate beautiful, self-contained HTML pages that visually explain systems, code changes, plans, and data. Use when the user asks for a diagram, architectur...
安全使用建议
This skill appears to do what it says (generate styled HTML reviews and diagrams), but review these points before installing:
- The prompts expect to run git and gh commands and to read/write files under your home directory (e.g., ~/.agent/diagrams/, ~/.agent/memory/, ~/.pi/agent/memory/). The registry entry did not declare these config/path dependencies — confirm you are comfortable with the skill reading project files and agent memory.
- Optional image generation uses surf-cli/Gemini if present; that path may require API keys or external network access. If you don't want external image calls, ensure surf-cli is not installed or restrict the agent's network access.
- There is a small Python script included (scripts/generate_header.py). Inspect any bundled scripts before executing them. Because there's no install step, the agent might still execute local code depending on how your agent platform runs skills — verify the platform's execution model.
- If you plan to use diff-review/plan-review/fact-check, test the skill in a sandbox repository first (no secrets, small sample repo) to confirm behavior.
- If you need assurance: ask the skill author for a short list of required binaries (git, gh, surf), explicit config paths the skill reads/writes, and a confirmation of what (if anything) is sent externally when surf is invoked. If unsure, decline installation or restrict the agent's file and network permissions.
功能分析
Type: OpenClaw Skill
Name: visual-explainer-main
Version: 1.0.0
The visual-explainer skill bundle is a legitimate tool designed to generate technical diagrams and HTML reports from codebase analysis. It gathers data using standard shell commands (git, grep, wc) and renders visualizations using reputable CDN-hosted libraries like Mermaid.js and Chart.js. The Python script (generate_header.py) uses proper HTML escaping, and the prompt templates are focused on technical accuracy and visual hierarchy without any evidence of data exfiltration, persistence mechanisms, or malicious prompt injection.
能力评估
Purpose & Capability
The skill's name and description align with generating HTML diagrams and visual reviews. However, SKILL.md and the prompt templates repeatedly assume access to git, the GitHub CLI (`gh`), the surf-cli/Gemini image pipeline, and to write/read paths under the user's home (e.g., ~/.agent/diagrams/, ~/.agent/memory/, ~/.pi/agent/memory/). The registry metadata declared no required binaries or config paths, so the skill's declared requirements understate what it actually expects to use.
Instruction Scope
Runtime instructions direct the agent to run many local shell commands (git diff, git show, grep, wc, gh pr diff, which surf), read entire changed files, and probe agent memory files (~/.agent/memory/*, ~/.pi/agent/memory/*). They also instruct writing files to ~/.agent/diagrams/ and opening them in a browser. Reading agent memory and arbitrary project files is powerful and privacy-sensitive; those accesses are within the stated purpose (code/plan/diff review) but are not explicitly declared in the skill metadata and merit user awareness.
Install Mechanism
There is no install spec (instruction-only) which is lower-risk, but the package includes a Python script (scripts/generate_header.py). The SKILL.md doesn't instruct running that script, but an agent or user could execute it — review the script before running. No external archives or untrusted downloads are used.
Credentials
The skill declares no required environment variables or credentials, yet its optional image generation path (surf-cli + Gemini) and the use of `gh` may implicitly require credentials or network access. The prompts also reference various home-dir paths (agent memory) that grant access to potentially sensitive conversation history — this level of local access should be explicitly documented and authorized by the user.
Persistence & Privilege
The skill does not request 'always: true' and does not modify other skills. It writes output to ~/.agent/diagrams/ and opens files in the browser — expected for a renderer. Autonomous invocation is allowed (platform default) but not combined here with other high-risk flags.
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 visual-explainer-main - 安装完成后，直接呼叫该 Skill 的名称或使用
/visual-explainer-main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
v1.0.0
Initial release of visual-explainer: generate styled, self-contained HTML pages for technical visualizations and data tables.
- Creates HTML diagrams and tables instead of ASCII art for complex or tabular technical content.
- Offers a range of visual aesthetics (blueprint, neon, editorial, hand-drawn, and more).
- Automatically generates HTML pages for tables with 4+ rows or 3+ columns.
- Integrates with Mermaid and Chart.js for various diagram types and dashboard layouts.
- Adds zoom controls to Mermaid diagrams by default for usability.
- Supports optional AI-generated images via surf-cli for enhanced visuals.
- Includes detailed workflow and styling guidance to ensure clear, varied, and accessible outputs.
元数据
常见问题
Visual Explainer Main 是什么？
Generate beautiful, self-contained HTML pages that visually explain systems, code changes, plans, and data. Use when the user asks for a diagram, architectur... 它是一个面向 Claude Code / OpenClaw 的 AI Agent Skill 插件，目前累计下载 347 次。
如何安装 Visual Explainer Main？
在 OpenClaw 或 Claude Code 对话框中运行命令「/install visual-explainer-main」即可一键安装，无需额外配置。
Visual Explainer Main 是免费的吗？
是的，Visual Explainer Main 完全免费（开源免费），可自由下载、安装和使用。
Visual Explainer Main 支持哪些平台？
Visual Explainer Main 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 Visual Explainer Main？
由 austindixson（@austindixson）开发并维护，当前版本 v1.0.0。
推荐 Skills