Generate responsive HTML pages suitable for reporting, supporting resizing and screenshot capture.

Generates a structured report HTML based on a specific template. Invoke when user wants to create a report, slide, or summary card from raw content.

Security Scan

The skill's declared purpose (generate responsive HTML reports) matches the provided instructions and template, but the shipped script is truncated in the package preview and imports subprocess/mentions image conversion — leaving a gap that prevents a full safety audit.

Purpose & Capability

Name, description, SKILL.md, and the visible portion of scripts/generate.py all align: the skill accepts structured JSON, renders an HTML template, and writes output to a workspace path. There are no unrelated declared env vars, binaries, or config paths.

Instruction Scope

SKILL.md instructs the agent to run the script with JSON and then use the browser tool to open file://<html_path> and take a screenshot — that stays within the stated purpose. However, the script's docstring claims it 'converts it to an image' and the script imports subprocess, indicating it may run external commands; the file preview is truncated so it's not possible to confirm whether the script itself shells out, calls remote services, or performs additional file/system access beyond writing the HTML.

Install Mechanism

No install spec; instruction-only with a Python script included. This minimizes install-time risk (nothing is downloaded at install).

Credentials

The skill declares no required environment variables, no credentials, and no config paths. That is proportional to the reported functionality.

Persistence & Privilege

Defaults are used (always: false, model invocation allowed). The skill does not request persistent platform privileges in the metadata.

What to consider before installing

The skill appears to do exactly what it says (generate an HTML report), but two things warrant caution before installing or running it in a production environment: (1) the provided preview of scripts/generate.py is truncated — the ending showing CLI parsing and any subprocess calls is missing, so you cannot fully audit what the script will run on your machine; (2) the script imports subprocess and the docstring mentions image conversion, so it may invoke external commands (e.g., headless browser, imagemagick, or curl) or shell out with user-provided data which could allow command injection or network activity. Before installing or executing: - Request the complete scripts/generate.py and review the remainder of the code for any subprocess.run / os.system / network calls, and confirm they don't pass unsanitized user input into shell commands. - If the script does call external binaries, prefer a sandboxed environment (container or VM) and run with least privilege. - If you plan to provide untrusted content, ensure the script properly sanitizes/escapes content to avoid HTML/JS injection or accidental execution when opened in a browser. - If you cannot obtain and review the full script, treat the skill as untrusted and avoid running it on sensitive hosts or with privileged credentials.

Report Generator Skill

This skill generates a beautifully formatted HTML report based on a clean, minimal business template style (Light & Concise).

Capabilities

1. Refine Content: Structures raw text into a 4-quadrant format (Goal, Q1-Q4 details, Summary).
2. Generate HTML: Creates a responsive HTML file.
3. Screenshot: (Instruction) Guides the agent to convert the HTML to an image using the browser tool.

Usage

1. Prepare Data

You need to extract the following information from the user's input:

• title: Report Title
• goal: One-sentence goal (Top section)
• q1 to q4: Four main sections, each containing:
  • title: Short title (e.g., "稳交付")
  • subtitle: Explanatory subtitle
  • slogan: Catchy slogan
  • items: List of bullet points (Array of strings)
• summary: Array of 4 keywords for the bottom summary

2. Run Generation Script

Execute the python script with the JSON data:

python3 scripts/generate.py --output "workspace/reports" --data '{"title": "...", ...}'

3. Convert to Image (Agent Action)

After the script returns the html_path, use the browser tool to screenshot it:

1. Open the file: file://<html_path>
2. Take a screenshot of the full page.

Example Data Structure

{
  "title": "2024 Q3 述职报告",
  "goal": "构建可规模化的非平台件能力体系",
  "q1": {
    "title": "稳交付",
    "subtitle": "解决扛量问题",
    "slogan": "保交付，争第一",
    "items": ["保障核心需求上线", "降低事故概率"]
  },
  "q2": { ... },
  "q3": { ... },
  "q4": { ... },
  "summary": ["稳定性", "增长", "效率", "反馈"]
}