Openclawx web-summarizer

Reads the content of an article or webpage, extracts the main points, and provides a concise, structured summary.

install

source · Clone the upstream repo

git clone https://github.com/next-open-ai/openclawx

Claude Code · Install into ~/.claude/skills/

T=$(mktemp -d) && git clone --depth=1 https://github.com/next-open-ai/openclawx "$T" && mkdir -p ~/.claude/skills && cp -r "$T/presets/workspaces/morning-briefing/skills/web-summarizer" ~/.claude/skills/next-open-ai-openclawx-web-summarizer && rm -rf "$T"

OpenClaw · Install into ~/.openclaw/skills/

T=$(mktemp -d) && git clone --depth=1 https://github.com/next-open-ai/openclawx "$T" && mkdir -p ~/.openclaw/skills && cp -r "$T/presets/workspaces/morning-briefing/skills/web-summarizer" ~/.openclaw/skills/next-open-ai-openclawx-web-summarizer && rm -rf "$T"

manifest: presets/workspaces/morning-briefing/skills/web-summarizer/SKILL.md

source content

Web Summarizer Skill

Use this skill to extract the core message from long articles, blog posts, or webpage content without the user having to read the whole text.

Workflow

Receive a URL from the user or from a previous step (like
```
news-fetcher
```
).
Use the
```
read_url_content
```
tool to retrieve the page content.
Analyze the returned markdown content, ignoring navigation elements, ads, and footers.
Generate a structured summary containing:
- Title: The main topic.
- TL;DR: A one-sentence summary.
- Key Points: 3 to 5 bullet points with the most important facts.
If the article is too long to read in one go, inform the user you are summarizing the beginning of the article.