lexmount-webfetch
Use Lexmount WebFetch through webfetch-cli for lightweight public page extraction and DOM dump tasks. Prefer this skill when the task needs structured extraction or rendered HTML capture without creating a live remote browser session.
ソース情報
- リポジトリ
- lexmount/webfetch-cli
- ソースの最終更新活動
- 2026年7月15日 08:17
- 検出された SKILL.md の言語
- 英語
- スター
- 0
- フォーク
- 0
インストール方法
デフォルトでは、最初にソースを確認する Prompt が選択されています。直接コマンドに切り替えるか、ローカルコピーをダウンロードすることもできます。
ソースファイルを確認
インストールを決める前に、SKILL.md と SkillsMP に表示されている付属ファイルをお読みください。
SKILL.md を表示中
SKILL.md
ソースの指示 · 読み取り専用プレビュー- name
- lexmount-webfetch
- description
- Use Lexmount WebFetch through webfetch-cli for lightweight public page extraction and DOM dump tasks. Prefer this skill when the task needs structured extraction or rendered HTML capture without creating a live remote browser session.
# Lexmount WebFetch CLI
Use `webfetch-cli` for Lexmount WebFetch extraction and DOM dump tasks.
## Fast Path
Do not run setup checks before every extraction. For ordinary fetch tasks, call the target command directly:
```bash
webfetch-cli extract --url <url>
webfetch-cli dump-dom --url <url>
```
Run checks only when they change the next action:
- First use in a new environment: `webfetch-cli --version` and `webfetch-cli auth status`.
- Missing credentials: `webfetch-cli auth login --open --connect-base-url https://browser.lexmount.cn`, then `webfetch-cli doctor --json`.
- Unexpected command/API failure: `webfetch-cli doctor --json`.
- Unsure installed capability or output shape: `webfetch-cli capabilities --json`.
- Skill file missing or stale: `webfetch-cli skill status`, then `webfetch-cli skill install --force` only if needed.
Never paste API keys into chat.
## Extract
```bash
webfetch-cli extract --url https://example.com
```
Default output is Markdown optimized for agents. It includes page metadata,
extraction quality warnings, and the extracted main text.
Use this for normal structured extraction. The CLI sends:
```json
{"extract":{"url":"https://example.com"}}
```
When debugging, request the full raw API response explicitly:
```bash
webfetch-cli extract --url https://example.com --include-trace --format json-full
webfetch-cli extract --url https://example.com --include-raw-dom --format json-full
```
Use `--format json` for compact structured data and `--format text` for plain text.
## Dump DOM
```bash
webfetch-cli dump-dom --url https://example.com
```
Default output is Markdown with DOM metadata, dump quality warnings, and captured
HTML. Debug fields are hidden unless `--format json-full` is used.
Use this when extraction needs rendered HTML or a reusable DOM snapshot. The CLI sends:
```json
{"url":"https://example.com"}
```
Use optional engine and cleanup hints when needed:
```bash
webfetch-cli dump-dom --url https://example.com --engine lightmount_dcl
webfetch-cli dump-dom --url https://example.com --filter-scripts-styles
```
GitHubで見る