xiaohongshu-image-generator
Analysis
The skill coherently generates local HTML card images and screenshots them, with only expected local server/browser use to be aware of.
Findings (2)
Artifact-based informational review of SKILL.md, metadata, install specs, static scan signals, and capability signals. ClawScan does not execute the skill or run runtime probes.
Checks for instructions or behavior that redirect the agent, misuse tools, execute unexpected code, cascade across systems, exploit user trust, or continue outside the intended task.
python3 -m http.server <端口号> ... browser action=open url=http://localhost:<端口号>/<filename>.html ... browser action=screenshot
The skill explicitly uses local command/server and browser screenshot tooling. This is expected for the stated image-generation purpose, but it gives the workflow access to whatever directory is served.
content_html = ''.join(f'<p>{line}</p>' for line in content_lines if line.strip()) ... html = TEMPLATE.format(title=config.get('title', '今日分享'), content_html=content_html or '<p>每天进步一点点</p>', tag=config.get('tag', '📌 今日分享'), author=config.get('author', '袁佳鹏')Prompt-derived text is inserted into generated HTML without visible escaping. Because the workflow renders that HTML in a browser, raw markup in inputs could be rendered as HTML rather than plain text.
