Awesome

About - v4.6

The idea behind waymore is to find even more links from the Wayback Machine than other existing tools.

👉 The biggest difference between waymore and other tools is that it can also download the archived responses for URLs on wayback machine so that you can then search these for even more links, developer comments, extra parameters, etc. etc. 👉 Also, other tools do not currenrtly deal with the rate limiting now in place by the sources, and will often just stop with incomplete results and not let you know they are incomplete.

Anyone who does bug bounty will have likely used the amazing waybackurls by @TomNomNoms. This tool gets URLs from web.archive.org and additional links (if any) from one of the index collections on index.commoncrawl.org. You would have also likely used the amazing gau by @hacker_ which also finds URL's from wayback archive, Common Crawl, but also from Alien Vault and URLScan. Now waymore gets URL's from ALL of those sources too (with ability to filter more to get what you want):

• Wayback Machine (web.archive.org)
• Common Crawl (index.commoncrawl.org)
• Alien Vault OTX (otx.alienvault.com)
• URLScan (urlscan.io)
• Virus Total (virustotal.com)

👉 It's a point that many seem to miss, so I'll just add it again :) ... The biggest difference between waymore and other tools is that it can also download the archived responses for URLs on wayback machine so that you can then search these for even more links, developer comments, extra parameters, etc. etc.

👉 PLEASE READ ALL OF THE INFORMATION ON THIS PAGE TO MAKE THE MOST OF THIS TOOL, AND ESPECIALLY BEFORE RAISING ANY ISSUES 🤘

👉 THIS TOOL CAN BE VERY SLOW, BUT IT IS MEANT FOR COVERAGE, NOT SPEED

⚠️ A common mistake that is made is passing a file of subdomains to get everything for a domain. DON'T DO IT! Just pass the domain only to get all subs for that domain. It will be SO much quicker, and you won't miss anything.

Installation

NOTE: If you already have a config.yml file, it will not be overwritten. The file config.yml.NEW will be created in the same directory. If you need the new config, remove config.yml and rename config.yml.NEW back to config.yml.

waymore supports Python 3.

Install waymore in default (global) python environment.

pip install waymore

OR

pip install git+https://github.com/xnl-h4ck3r/waymore.git -v

You can upgrade with

pip install --upgrade waymore

pipx

Quick setup in isolated python environment using pipx

pipx install git+https://github.com/xnl-h4ck3r/waymore.git

Usage

Run with docker

Install docker

git clone https://github.com/xnl-h4ck3r/waymore.git
cd waymore

Build image:

docker build -t waymore .

Run waymore with this command:

docker run -it --rm -v $PWD/results:/app/results waymore:latest waymore -i example.com -oU example.com.links -oR results/example.com/

Input and Mode

The input -i can either be a domain only, e.g. redbull.com or a specific domain and path, e.g. redbull.com/robots.txt. You can also pass a file of domains/URLs to process (or pass values in by piping from another program on the command line). NOTE: Any scheme, port number, query string, or URL fragment will be removed from the input values. However, if you provide a path, this will be specifically searched for, so will limit your results.

There are different modes that can be run for waymore. The -mode argument can be 3 different value:

• U - URLs will be retrieved from archive.org (if -xwm is not passed), commoncrawl.org (if -xcc is not passed), otx.alienvault.com (if -xvv is not passed) and urlscan.io (if -xus is not passed)
• R - Responses will be downloaded from archive.org
• B - Both URLs and Responses will be retrieved

If the input was a specific URL, e.g. redbull.com/robots.txt then the -mode defaults to R. Only responses will be downloaded. You cannot change the mode to U or B for a domain with path because it isn't necessary to retrieve URLs for a specific URL.

If the input is just a domain, e.g. redbull.com then the -mode defaults to B. It can be changed to U or R if required. When a domain only is passed then all URLs/responses are retrieved for that domain (and sub domains unless -n is passed). If the no sub domain option -n is passed then the www sub domain is still included by default.

config.yml

The config.yml file (typically in ~/.config/waymore/) have values that can be updated to suit your needs. Filters are all provided as comma separated lists:

• FILTER_CODE - Exclusions used to exclude responses we will try to get from web.archive.org, and also for file names when -i is a directory, e.g. 301,302. This can be overridden with the -fc argument. Passing the -mc (to match status codes instead of filter) will override any value in FILTER_CODE or -fc.

• FILTER_MIME - MIME Content-Type exclusions used to filter links and responses from web.archive.org through their API, e.g. 'text/css,image/jpeg. This can be overridden with the -ft argument. . Passing the -mt (to match MIME types instead of filter) will override any value in FILTER_MIME or -ft.

• FILTER_URL - Response code exclusions we will use to filter links and responses from web.archive.org through their API, e.g. .css,.jpg

• FILTER_KEYWORDS - Only links and responses will be returned that contain the specified keywords if the -ko/--keywords-only argument is passed (without providing an explicit value on the command line), e.g. admin,portal

• URLSCAN_API_KEY - You can sign up to urlscan.io to get a FREE API key (there are also paid subscriptions available). It is recommended you get a key and put it into the config file so that you can get more back (and quicker) from their API. NOTE: You will get rate limited unless you have a full paid subscription.

• CONTINUE_RESPONSES_IF_PIPED - If retrieving archive responses doesn't complete, you will be prompted next time whether you want to continue with the previous run. However, if stdout is piped to another process it is assumed you don't want to have an interactive prompt. A value of True (default) will determine assure the previous run will be continued. if you want a fresh run every time then set to False.

• WEBHOOK_DISCORD - If the --notify-discord argument is passed, knoxnl will send a notification to this Discord wehook when a successful XSS is found.

• DEFAULT_OUTPUT_DIR - This is the default location of any output files written if the -oU and -oR arguments are not used. If the value of this key is blank, then it will default to the location of the config.yml file.

  NOTE: The MIME types cannot be filtered for Alien Vault OTX and Virus Total because they don't have the ability to filter on MIME Type. Sometimes URLScan does not have a MIME Type defined for a URL. In these cases, URLs will be included regardless of filter or match. Bear this in mind and consider excluding certain providers if this is important.

Output

In the default output directory specificed in the config.yml file with DEFAULT_OUTPUT_DIR (if that is blank, it will default to the location of the config.yml file itself, typically ~/.config/waymore/), a results directory will be created. Within that, a directory will be created with target domain (or domain with path) passed with -i (or for each line of a file passed with -i). You can alternatively use argument -oU to specify where the URL links file will be output (and the name of the file). You can also use argument -oR to specify a directory (or path) where the archived responses will be output. When run, the following files are created in the target directory:

• waymore.txt - If -mode is U or B, this file will contain links from selected sources. Links will be retrieved from archive.org Wayback Machine (unless -xwm was passed), commoncrawl.org (unless -xcc was passed), otx.alienvault.com (unless -xav was passed) and urlscan.io (unless -xus was passed). If the -ow option was also passed, any existing waymore.txt file in the target results directory will be overwritten, otherwise new links will be appended and duplicates removed.
• index.txt - If -mode is R or B, and -url-filname was not passed then archived responses will be downloaded and hash values will be used for the saved file names. This file contains a comma separated list of <hash>,<archive URL>,<timestamp> in case you need to know which URLs produced which response.
• ALL OTHER FILES - These archived response files will be created if -mode was R or B. If -url-filename was passed the the file names will be the archive URL that generated the response, e.g. https--example.com-robots.txt, otherwise the file name will be a hash value, e.g. 7960113391501.{EXT} where {EXT} will be the extension derived from the path, otherwise it will be derived from the response content-type. Sometimes a general extesions will be given, e.g. .js for any content type containing the word javascript, otherwise it may be set to the last part of the type (e.g. if it's application/java-archive the file will be 7960113391501.java-archive). Using hash values mean that less files will be written as there will only be one file per unique response. These archived responses are edited, before being saved, to remove any reference to web.archive.org.

Info and Suggestions

The number of links found, and then potentially the number of files archived responses you will download could potentially be HUGE for many domains. This tool isn't about speed, it's about finding more, so be patient.

There is a -p option to increase the number of processes (used when retrieving links from all sources, and downloading archived responses from archive.org). However, although it may not be as fast as you'd like, I would suggest leaving -p at the default of 3 because I personally found issues with getting responses with higher values. We don't want to cause these services any problems, so be sensible!

I often use the -f option because I want waymore.txt to contain ALL possible links. Even though I don't care about images, fonts, etc. it could still be useful to see all possible paths and maybe parameters. Any filters will always be applied to downloading archived responses though. You don't want to waste time downloading thousands of images!

Using the -v option can help see what is happening behind the scenes and could help you if you aren't getting the output you are expecting.

All the MIME Content Types of URL's found (by all sources except Alien Vault) will be displayed when -v is used. This may help to add further exclusions if you find you still get back things you don't want to see. If you spot a MIME type that is being included but you don't want that going forward, add it to the FILTER_MIME in config.yml. It should be noted that sometimes the MIME type on archive.org is stored as unk and unknown instead of the real MIME so the filter won't necessarily remove it from their results. The FILTER_URL config settings can be used to remove these afterwards. For example, if a GIF has MIME type unk instead of image/gif (and that's in FILTER_MIME) then it won't get filtered, but if the url is https://target.com/assets/logo.gif and .gif is in FILTER_URL it won't get requested.

If config.yml doesn't exist, or the entries for filters, aren't in the file, then default filters are used. It's better to have the file and review these to ensure you are getting what you need.

There can potentially be millions of responses so make sure you set filters, but also the Limit (-l), From Date (-from), To Date (-to) and/or Capture Interval (-ci) if you need to. The limit defaults to 5000, but say you wanted to get the latest 20,000 responses from 2015 up until January 2018... you would pass -l -20000 -from 2015 -to 201801. The Capture Interval determines how many responses will get downloaded for a particular URL within a specified period, e.g. if you set to m you will only get one response per month for a URL. The default d will likely greatly reduce the number of responses and unlikely to miss many unique responses unless a target changed something more than once in a given day.

Another useful argument is -mc that will only get results where the HTTP status code matches the comma separated list passed, e.g. -mc 200 or -mc 200,403.

You can also greatly reduce the number (and therefore reduce the execution time) of links and responses by only returning ones that contain keywords you are interested in. You can list these keywords in config.yml with the FILTER_KEYWORDS key and then pass argument -ko/--keywords-only to use these, or you can pass -ko/--keywords-only with a specific Regex, e.g. -ko "\.js(\?|$)" to only get JS files.

As mentioned above, sign up to urlscan.io to get a FREE API key (there are also paid subscriptions available). It is recommended you get a key and put it into the config.yml file so that you can get more back (and quicker) from their API. NOTE: You will get rate limited unless you have a full paid subscription.

The archive.org Wayback Machine CDX API can sometimes can sometimes require a huge amount of requests to get all the links. For example, if you run waymore for -i twitter.com it says there are 28,903,799 requests to archive.org that need to be made (that could take almost 1000 days for some people!!!). The argument -lr can be used to limit the number of requests made per source (although it's usually archive.org that is the problem). The default value for the argument is 0 (Zero) which will apply no limit.

There is also a problem with the Wayback Machine CDX API where the number of pages returned is not correct when filters are applied and can cause issues (see https://github.com/internetarchive/wayback/issues/243). Until that issue is resolved, setting the -lr argument to a sensible value can help with that problem in the short term.

The Common Crawl API has had a lot of issues for a long time. Including this source could make waymore take a lot longer to run and may not yield any extra results. You can check if tere is an issue by visiting http://index.commoncrawl.org/collinfo.json and seeing if this is successful. Consider excluding Common Crawl altogether using the --providers argument and not including commoncrawl, or using the -xcc argument.

The provider API servers aren't designed to cope with huge volumes, so be sensible and considerate about what you hit them with!

When downloading archived responses, this can take a long time and can sometimes be killed by the machine for some reason, or manually killed by the user. In the targets results directory, a file called responses.tmp is created at the start of the process and contains all the response URLs that will be retrieved. There will also be a file called continueResp.tmp that stores the index of the latest response retrieved. If waymore is run to get responses (-mode R or -mode B), and these files exist, it means there was a previous incomplete run, and you will be asked if you want to continue with that one instead. It will then continue from where it stopped before.

Some Basic Examples

Example 1

Just get the URLs from all sources for redbull.com (-mode U is just for URLs, so no responses are downloaded):

The URLs are saved in the same path as config.yml (typically ~/.config/waymore) under results/redbull.com/waymore.txt

Example 2

Get ALL the URLs from Wayback for redbull.com (no filters are applied in mode U with -f, and no URLs are retrieved from Commone Crawl, Alien Vault, URLScan and Virus Total, because -xcc, -xav, -xus, -xvt are passed respectively. This can also be achieved by passing --providers wayback instead of the exclude arguments). Save the FIRST 200 responses that are found starting from 2022 (-l 200 -from 2022):

The -mode wasn't explicitly set so defaults to B (Both - retrieve URLs AND download responses). A file will be created for each unique response and also saved in results/redbull.com/:

There will also be a file results/redbull.com/index.txt that will contain a reference to what URLs gave the response for what file, e.g.

4847147712618,https://web.archive.org/web/20220426044405/https://www.redbull.com/additional-services/geo ,2022-06-24 20:07:50.603486

where 4847147712618 is the hash value of the response in 4847147712618.xnl, the 2nd value is the Wayback Machine URL where you can view the actual page that was archived, and the 3rd is a time stamp of when the response was downloaded.

Example 3

You can pipe waymore to other tools. Any errors are sent to stderr and any links found are sent to stdout. The output file is still created in addition to the links being piped to the next program. However, archived responses are not piped to the next program, but they are still written to files. For example:

waymore -i redbull.com -mode U | unfurl keys | sort -u

You can also pass the input through stdin instead of -i.

cat redbull_subs.txt | waymore

Example 4

Sometimes you may just want to check how many request, and how long waymore is likely to take if you ran it for a particular domain. You can do a quick check by using the -co/--check-only argument. For example:

waymore -i redbull.com --check-only

Finding Way More URLs!

So now you have lots of archived responses and you want to find extra links? Easy! Why not use xnLinkFinder? For example:

xnLinkFinder -i ~/Tools/waymore/results/redbull.com -sp https://www.redbull.com -sf redbull.com -o redbull.txt

Or run other tools such as trufflehog or gf over the directory of responses to find even more from the archived responses!

Issues

If you come across any problems at all, or have ideas for improvements, please feel free to raise an issue on Github. If there is a problem, it will be useful if you can provide the exact command you ran and a detailed description of the problem. If possible, run with -v to reproduce the problem and let me know about any error messages that are given.

TODO

• Add an -oss argument that accepts a file of Out Of Scope subdomains/URLs that will not be returned in the output, or have any responses downloaded

References

• Wayback CDX Server API - BETA
• Common Crawl Index Server
• Alien Vault OTX API
• URLScan API
• VirusTotal API (v2)

Good luck and good hunting! If you really love the tool (or any others), or they helped you find an awesome bounty, consider BUYING ME A COFFEE! ☕ (I could use the caffeine!)

🤘 /XNL-h4ck3r