| name | zai-web-reader |
| description | Fetch and extract content from any webpage by URL using Z.AI Web Reader. Use when the user provides a specific URL and wants its content read, summarized, or analyzed — API docs, articles, open-source project pages, blog posts. Run scripts/read.py via Bash. |
| compatibility | Requires Python 3 and Z_AI_API_KEY environment variable. |
| metadata | {"author":"z-ai","version":"1.0","api-docs":"https://docs.bigmodel.cn/cn/coding-plan/mcp/reader-mcp-server"} |
Z.AI Web Reader Skill
Read any webpage by running scripts/read.py with Bash. No MCP configuration needed — set Z_AI_API_KEY and call the script directly.
Prerequisites
export Z_AI_API_KEY=your_api_key
Usage
python3 scripts/read.py <url>
Returns: page title, body text, metadata, and links.
Examples
python3 scripts/read.py https://docs.python.org/3/library/asyncio.html
python3 scripts/read.py https://github.com/openai/openai-python
python3 scripts/read.py https://example.com/blog/post
When to Use
- User provides a specific URL to read or summarize
- User wants to read API documentation at a given URL
- User wants to understand a webpage or article
- User asks to "read this link", "summarize this page", or "check this URL"
vs. zai-web-search: Use this skill when you have a URL. Use zai-web-search when you need to find pages first.
vs. zai-zread: For GitHub repos, prefer zai-zread which understands repo structure, issues, and PRs.
Instructions for the Agent
- Extract the URL from the user's message.
- Run
python3 scripts/read.py <url> via Bash.
- Process the returned content and focus on the sections relevant to the user's question.
- Summarize and present the information, citing the source URL.
If the page is behind authentication or a paywall, the tool may not retrieve the full content — inform the user.
