AI Agent Hub
Back to skills
Prohibited Word Compliance Checker icon

Prohibited Word Compliance Checker

Business Operations Updated 2026.08.30

Paste the following prompt into your AI chat to install this skill:

Please install @user_4bfc9f62/content-check using the guide at https://skillhub.cn/install/skillhub.md.

About this skill

Problem

Before publishing, phrases such as 'best', 'works in 7 days', or 'add me on WeChat' may carry different risk on Xiaohongshu, Douyin, WeChat Official Accounts, Zhihu, Bilibili, Taobao, and JD. Manual review is slow and makes it hard to compare platform-specific rules, severity, and replacement options. This skill is useful for a quick pre-publish scan that produces a reviewable prohibited-word report.

How It Works

The skill works through two files: scripts/check.py for scanning and scripts/word_library.py with a built-in library of 1,393+ terms. It first checks whether --platform is set; if not, it scans the full platform library. It then filters terms by platform and returns matched words, risk level, and replacement suggestions. Risk levels include high, medium, and low, corresponding to required changes, recommended changes, and discretionary changes. Common parameters include --text for input text and --json for structured output, which can be used in scripts or content pipelines.

Boundaries

This skill is a compliance pre-check, not legal review or platform appeal support. Politically sensitive terms are flagged by risk only, without exposing specific entries. Replacement suggestions aim for semantic equivalence and should not be applied blindly. For batch checks, run the script per text segment and aggregate the reports to avoid losing context in very long inputs.

Use Cases

  • Before posting a Xiaohongshu note, scan copy for absolute claims or share-bait and export replacement suggestions.
  • When reviewing Douyin campaign copy, filter terms by platform to reduce false positives and label risk levels.
  • For a content team, check WeChat, Zhihu, and Bilibili drafts segment by segment and aggregate high and medium risks.
  • Before publishing product detail pages, flag phrases like guaranteed results or risk-free returns and generate compliant edits.

Best For

  • Content operators publishing on Xiaohongshu, Douyin, or WeChat who need quick pre-publish compliance checks.
  • Marketing specialists managing multi-channel templates who need platform-specific false positives and risk lists.
  • Engineers building content pipelines who need JSON detection reports for automated publishing or review tools.
  • Compliance reviewers checking e-commerce copy who need to flag absolute claims, false promises, and share-bait.