← Back to Skills Marketplace
215
Downloads
0
Stars
0
Active Installs
1
Versions
Install in OpenClaw
/install dead-link-scanner
Description
Scan websites, markdown files, and HTML files for broken links (dead links). Use when checking a website for 404s, validating links in documentation or READM...
README (SKILL.md)
dead-link-scanner
Find broken links in websites, markdown files, and HTML documents.
Quick Start
# Scan a website for broken links
python3 scripts/dead_link_scanner.py scan https://example.com
# Scan with depth limit (default: 1)
python3 scripts/dead_link_scanner.py scan https://example.com --depth 3
# Scan a local markdown file
python3 scripts/dead_link_scanner.py file README.md
# Scan multiple files
python3 scripts/dead_link_scanner.py file docs/*.md
# JSON output
python3 scripts/dead_link_scanner.py scan https://example.com --json
# Only show broken links
python3 scripts/dead_link_scanner.py scan https://example.com --broken-only
Commands
scan
Crawl a website and check all links on each page.
python3 scripts/dead_link_scanner.py scan \x3Curl> [options]
Options:
--depth \x3Cn>— Max crawl depth (default: 1, 0 = single page only)--timeout \x3Cseconds>— Request timeout (default: 10)--json— Output results as JSON--broken-only— Only show broken links--internal-only— Only check links within the same domain--max-urls \x3Cn>— Max URLs to check (default: 200)--delay \x3Cseconds>— Delay between requests (default: 0.2)
file
Scan local markdown or HTML files for broken links.
python3 scripts/dead_link_scanner.py file \x3Cpath>... [options]
Options:
--timeout \x3Cseconds>— Request timeout (default: 10)--json— Output as JSON--broken-only— Only show broken links
Output
Default text output:
✓ 200 https://example.com/about
✓ 200 https://example.com/blog
✗ 404 https://example.com/old-page (found on: https://example.com)
✗ ERR https://dead-domain.xyz (found on: https://example.com) — ConnectionError
Summary line:
Checked 42 links: 39 OK, 3 broken
Usage Guidance
This skill is coherent and implements a dead-link scanner in Python. Before installing or running it, be aware that: (1) it will make outbound HTTP(S) requests to every discovered URL (including internal/intranet addresses you point it at), which can trigger server logs or expose your agent's IP; (2) it reads only files you pass on the command line — avoid scanning sensitive files you don't intend to expose; (3) it does not appear to obey robots.txt and will crawl pages according to the provided depth/max-urls/delay settings, so use polite settings (delay, max-urls) when crawling third-party sites; and (4) the User-Agent includes an identifying URL, so contacted servers will see that string. If you need autonomous scanning across private networks or sensitive repos, review the script and enforce access controls and network restrictions first.
Capability Analysis
Type: OpenClaw Skill
Name: dead-link-scanner
Version: 1.0.0
The 'dead-link-scanner' skill is a legitimate utility for identifying broken links in websites and local files. The Python script (scripts/dead_link_scanner.py) uses standard libraries (urllib, html.parser) to crawl URLs and parse files, incorporating safety features such as request delays, depth limits, and a maximum URL count to prevent abuse. No evidence of data exfiltration, malicious execution, or prompt injection was found.
Capability Assessment
Purpose & Capability
Name/description match the included Python implementation: the code crawls HTML pages, extracts links from HTML and Markdown, and checks links' HTTP status. It does not request unrelated capabilities (no cloud creds, no extra binaries).
Instruction Scope
SKILL.md instructs the agent to run the included script to scan a URL or local files. The script only reads files explicitly passed to the 'file' command and performs network requests for discovered links; it does not attempt to read arbitrary system state or extra environment variables.
Install Mechanism
There is no install spec; this is instruction+code only. The Python script is bundled with the skill and nothing is downloaded or executed from external installers.
Credentials
The skill requires no environment variables, credentials, or config paths. The single notable hard-coded value is a USER_AGENT string (dead-link-scanner/1.0 (+https://clawhub.com)), which is benign but reveals an identifying User-Agent to servers it contacts.
Persistence & Privilege
Skill does not request persistent or elevated privileges. always is false and it does not modify other skills or system configurations. It can be invoked autonomously by the agent (default), which is expected behavior for a skill.
How to Use
- Make sure OpenClaw is installed (local or Docker)
- Run the install command in chat:
/install dead-link-scanner - After installation, invoke the skill by name or use
/dead-link-scanner - Provide required inputs per the skill's parameter spec and get structured output
Version History
v1.0.0
Initial release — scan websites and files for broken links.
- Scan websites recursively with configurable depth and options.
- Check local markdown and HTML files for dead links.
- Multiple output formats: plain text or JSON.
- Filter results to show only broken or internal links.
- Customizable crawl parameters, including timeouts, delays, and URL count limits.
Metadata
Frequently Asked Questions
What is Dead Link Scanner?
Scan websites, markdown files, and HTML files for broken links (dead links). Use when checking a website for 404s, validating links in documentation or READM... It is an AI Agent Skill for Claude Code / OpenClaw, with 215 downloads so far.
How do I install Dead Link Scanner?
Run "/install dead-link-scanner" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.
Is Dead Link Scanner free?
Yes, Dead Link Scanner is completely free, licensed under MIT-0. You can download, install and use it at no cost.
Which platforms does Dead Link Scanner support?
Dead Link Scanner is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).
Who created Dead Link Scanner?
It is built and maintained by John Wang (@johnnywang2001); the current version is v1.0.0.
More Skills