Usage

Synopsis

link_check [OPTIONS] [ROOT_URL]

ROOT_URL is the URL to start crawling from. It may also be specified in a configuration file.

Options

Flag	Type	Default	Description
ROOT_URL	string	required unless set in config	Root URL to begin crawling when not provided by --config-file.
-o, --output	path	stdout	File path for the final plain-text report.
--log-file	path	stderr	File path for log messages.
--log-level	string	INFO	Minimum log level: DEBUG, INFO, WARNING, ERROR, CRITICAL. Case-insensitive (e.g. debug and DEBUG both work).
--timeout	int	10	Timeout in seconds for each HTTP request.
--retries	int	3	Number of retry attempts for transient failures.
--max-requests	int	unlimited	Maximum total HTTP requests to issue.
--max-depth	int	unlimited	Maximum directory depth to crawl.
--max-threads	int	10	Maximum number of concurrent threads.
--max-referencing-pages	int	10	Max referencing pages per URL in report.
--config-file	path	none	Path to a YAML configuration file.
--version	flag	—	Print version and exit.

Examples

Check a website with default settings:

link_check https://example.com

Save report to a file, limit depth and threads:

link_check https://example.com --max-depth 3 --max-threads 20 -o report.txt

Use a configuration file:

link_check --config-file config.yaml

Override config file timeout on the command line:

link_check --config-file config.yaml --timeout 30

Enable debug logging to a file:

link_check https://example.com --log-level debug --log-file crawl.log

Progress Reporting

While a crawl is running, a progress line is printed to stderr every 5 seconds:

[Progress] 365/~410 URLs checked | 12.2 URLs/s | 45 in queue | 8 threads active | 0m 30s elapsed

The fields are:

• checked/~estimate — URLs fully processed so far, and an estimate of the total (checked + currently queued).

• URLs/s — average request rate since the crawl started.

• in queue — URLs waiting to be submitted to the thread pool.

• threads active — worker threads currently executing (bounded by --max-threads).

• elapsed — wall-clock time since the crawl started.

Interrupting a Crawl

Press Ctrl-C once to abort. Any in-flight HTTP requests are allowed to finish, then a partial report is generated from results collected so far and written to the configured output destination. A second Ctrl-C issues an immediate hard kill.

Exit Codes

Code	Meaning
0	All checks passed: no broken links, no non-200 responses, no broken anchors, no misplaced assets, no SSL errors.
1	One or more problems detected.
2	Fatal error: invalid arguments, config file not found or has unknown keys, etc.
130	Crawl was interrupted with Ctrl-C; report contains partial results.