xiaohongshu card generator

AdvisoryAudited by Static analysis on May 10, 2026.

Overview

Detected: suspicious.dangerous_exec

Findings (1)

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.

What this means

Rendering an MDX file from someone else could run code in the local rendering process.

Why it was flagged

When --mdx-mode is enabled, the input Markdown/MDX file is passed to @mdx-js/mdx evaluate(), which can run embedded JavaScript during rendering rather than treating the document as inert text.

Skill content
const module = (await evaluate(markdown, { ... })) ... if (options.mdxMode) { return { html: await mdxToHtml(markdown), warnings }; }
Recommendation

Only use --mdx-mode with trusted files, or disable/sandbox MDX evaluation and clearly warn users before rendering untrusted MDX.

What this means

A malicious Markdown file could affect the browser render step or make unexpected network requests instead of being rendered as plain content.

Why it was flagged

Normal Markdown mode explicitly preserves raw HTML instead of escaping or sanitizing it, so embedded HTML such as scripts, event handlers, or remote resources can remain active when the generated page is rendered.

Skill content
.use(remarkRehype, { allowDangerousHtml: true }).use(rehypeRaw).use(rehypeStringify, { allowDangerousHtml: true })
Recommendation

Sanitize raw HTML by default, provide a trusted-content mode for advanced HTML, and render untrusted documents in a restricted no-network sandbox.

What this means

Installing the skill will add local Node packages and a browser runtime.

Why it was flagged

The setup script installs npm dependencies and downloads Playwright Chromium. That is normal for this renderer, but it is an external dependency/download surface.

Skill content
npm --prefix "${SCRIPT_DIR}" ci ... npx --prefix "${SCRIPT_DIR}" playwright install chromium
Recommendation

Run setup only from a trusted source, avoid elevated privileges, and review package.json/package-lock.json if supply-chain risk matters.

Findings (1)

critical

suspicious.dangerous_exec

Location
scripts/xhs-card.cjs:8
Finding
Shell command execution detected (child_process).