Wechat Image Generator

Generate beautiful images for WeChat articles (covers, comparisons, charts) with zero token cost. Use when user needs images for social media posts, article...

Security Scan

The skill's code, templates, and runtime instructions align with its stated purpose of generating HTML-based images and using a local/browser screenshot tool; no unrelated credentials, network exfiltration, or surprising installs are requested — though a small implementation bug and a local command-injection risk when running an auxiliary script are worth noting.

Purpose & Capability

Name/description match the included files: Python scripts + HTML templates that produce cover/compare/chart HTML and guidance to capture screenshots via the OpenClaw 'browser' tool. Required binary (python3) and requested browser tool are appropriate and proportional. No unrelated environment variables, services, or install steps are present.

Instruction Scope

SKILL.md and README restrict actions to generating HTML files and opening them in a browser for screenshots using the OpenClaw browser tool or a local HTTP server. They do not instruct reading unrelated files or sending data to external endpoints. Two implementation issues to note: (1) generate.py prints the 'browser' commands for the user/agent to run rather than executing them directly (this is expected/safer), and (2) scripts/auto_screenshot.py attempts to invoke the 'browser' commands via subprocess.run(shell=True) but does not actually execute the final combined command (the code prints the combined_cmd and returns True). Additionally, auto_screenshot.py constructs shell commands with user-supplied file paths and executes them with shell=True — this creates a potential local command-injection risk if untrusted paths are passed to that script. This is an implementation vulnerability, not evidence of network exfiltration or unrelated scope creep.

Install Mechanism

No install spec or remote downloads. All code and templates are included in the package; there is no external installation or archive extraction. This is low-risk from an install mechanism perspective.

Credentials

The skill requires no environment variables or credentials. The sole declared runtime dependency (python3) and the implicit dependency on an OpenClaw 'browser' tool are directly relevant to the stated functionality.

Persistence & Privilege

The skill does not request always:true, does not modify other skills, and does not require system-level configuration. It is user-invocable and will not be force-included or persistently privileged.

Assessment

This skill appears to do what it says: generate static HTML templates and instruct you to take screenshots via a local browser tool. Before installing/using: (1) review and prefer running generate.py (which only writes HTML and prints safe commands) rather than auto_screenshot.py; the latter executes shell commands with shell=True and could be abused if you pass untrusted file paths. (2) The auto-screenshot behavior is partly unimplemented/buggy — screenshots are typically taken by running the printed 'browser open ... && browser screenshot ...' command yourself or using the serve.py + local http.server flow. (3) Run the scripts in a trusted or sandboxed environment if you are unsure about the 'browser' tool implementation. (4) If you plan to integrate this into automation, modify subprocess usage to avoid shell=True and to validate/escape paths to eliminate command-injection risk. Overall the package is coherent with its purpose but exercise standard caution when running helper scripts that invoke shell commands.

Generate beautiful images for WeChat articles with zero token cost and auto-screenshot.

Quick Start

1. Cover Image (封面图)

python3 scripts/generate.py cover \
  --title "我的第一个开源项目" \
  --subtitle "Token 成本降低 90%" \
  --output output/cover.png

2. Comparison Image (对比图)

python3 scripts/generate.py compare \
  --left "# Markdown\n**Bold** text" \
  --right "HTML 渲染结果" \
  --label "1 秒转换" \
  --output output/compare.png

3. Chart Image (数据图)

python3 scripts/generate.py chart \
  --data "Token消耗:8000,650|生成耗时:20,1" \
  --labels "AI生成,预制模板" \
  --output output/chart.png

Workflow

1. Run generator script with parameters
2. Script creates HTML with embedded data
3. Opens in browser via OpenClaw browser tool
4. Auto-screenshot and save to output folder

Design Philosophy

Zero token cost: All templates are pre-built HTML/CSS Auto-screenshot: Integrated with OpenClaw browser tool Customizable: Easy to modify templates for different styles

Token Cost Analysis

Per image generation:

• Read SKILL.md: ~500 tokens (first time only)
• Execute script: ~100 tokens
• Browser screenshot: ~50 tokens

Total: ~650 tokens vs DALL-E/Midjourney ~1000-5000 tokens per image.

Requirements

• Python 3
• OpenClaw browser tool (for auto-screenshot)