--- This project is depreciated as fb made major changes ---
Somedo Version 0.6.1 Alpha 2019-02-03
1 Introduction
Somedo is a downloader for social media platforms such as Facebook. The downloaded sites/profiles are mainly stored as screenshots (PNG) and PDF files. Additional data can also be found in other files such as CSV, JSON, HTML etc.
Somedo is designed for open source intelligence and law enforcement but it might be helpful in other fileds too. Somedo has a modular structure. Modules for Facebook, Instagram an Twitter are implemented so far.
Somedo ist in alpha state and might never leave beta due to constant development of the social media platforms.
There are socialmedia platforms that require an account for the investigator to see anything. Facebook is one of them. Register one or more accounts avoiding giving away any real personal data. Set the privacy level and options for the accounts to the maximum. The accepted accounts can then be used to optain data.
Make sure this is leagel for you in your country/state etc. I, the developer of this open source software, is and will never be responsable for any use of my tiny one man show development. Somedo is intended to protect the law and may help aquiring information for journalistic or other goals that do not interfere with legislation, privacy and human rights.
2 Use (GUI)
You need Googel Chrome or Chromium to be installed.
As the downloads might take a while the idea is to add jobs to a list and start the execution when you know what you want. The investigator can use the browser of his choice to detect the targeted account(s) or criterias to download.
The modules should by default take several accounts as targets. Seperate them with comma or semicolon in the target(s) field.
You can add jobs hitting the large Buttons for Facebook, Instagram or Twitter. To edit click the job in the list. The job list can be resetted usind "Purge jobs". Make shure that in Configuration is the correct path of Google's Chrome or the open version Chromium. You can change the file path (write it into the field or use the hand symbol button). In addition you might want to change to directory where the optained data will be stored by setting "Output directory".
Before launching make sure the used browser (Chrome/Chromium) is not running. Somedo will hide the Google browser (in VISIBLE mode you are able to see the browser working but no PDF files will be created due to limitations of the browser's abilities - info for developers only).
The jobs are execute by hitting "Start jobs". At this point the browser used by Somedo has to be closes (if you use Chromium for Somedo, you can still work with Chrome).
A running task can be aborted by the user hitting Stop running task (as the already obtained data is stored, it might take a while before All done is reported).
While running Somedo gives some information about what it is doing (Messages filed). In addition it might be a good idea to observer the progress by looking into the target directories.
Chrome/Chromium ist started headless so you will not see anything while Somedo is executing the given jobs. As Somedo does not use APIs but opens the pages as a human user would do, it does not work very fast. The job which is currently executed is shown by flashing of the list entry.
The login credentials can set in the Options field. These and the paths to Chrome/ Chromium and the output directory can be saved to file an loaded to avoid typing marathons.
You can save the configuration such as the output directory, path to Chrome/Chromium, user names, passwords etc. to a config file to load it after a new launch of Somedo using Load configuration.
3 Modules
3.1 Facebook
At least one valid Facebook account is needed to login. You can use several accounts. Especially with the Network option, it is likely that Facebook will block an account. Somedo then will switch to the next geven credentials. In the fields Email and Password you have to seperate email addresses (or phone numbers) are seperated by comma or semicolon (e.g. "[email protected], [email protected]" / "hecklerandkoch, beretta"). If all investigator accounts have the same password, it does not to be repeated.
Targets might look like this: "1000023456789; holly.gennero". It is recommended to use Facebook IDs. From complete URLs, Somedo tries to filter out the ID. The displayed Facebook name is not individual and does not indicate a certain account. Somedo is NOT a tool to search for an individual account.
Somedo might download data from your targeted accounts including befriendet or related accounts.
Limits are set as a lot of accounts are to large to get completely. A directory to store the obtained data (creenshots, PDF, CSV, HTML etc.) is proposed but can be set (klick on the hand symbol right to entry field. The screenshots are designed to fit on letter size or DIN A4. Data is only obtained on the number screenshotted pages so the max. number of screenshots also limits downloaded photos and the Extend Network option.
The Network option will try to get the Facebook Friends and on Extend Network accounts that left comments or likes. Network Depth 0 downloads the Friend lists of the targets. Depth 1 will should you a network visualisation. Look in the directory "Facebook/Network". "Facebook/Network/network.html" will show Facebook Friendships with solid lines and commentors/likers with dashed arrows. You can click on the nodes to get account infos. Depth 2 gets friends of friends and so on. Be aware of the exponential number of befriended accounts.
If target is a Page ("pg") or a Facebook Group ("groups"), Somedo tries to aquire data but this does not work very relieable. E.g. getting all members of large Facebook groups is impossible. The possible number of screenshots is not only limited by Chrome/Chromium but also by the Facebook servers. If there is no response from server Somedo will just abort and go on to the next task.
3.2 Instagram
Landing gets the head of an Instagrm Web page. Media downloads photos and Videos. The Max. number of screenshots will also limit the downloaded files (be aware that the max. number of screenshots is NOT the number of files to download)
3.3 Twitter
By default the targets are Twitter accounts. the Search option takes Target(s) as search pattern. You cannot use both in one job.
- CLI
The CLI does not work when you are using the EXE file. In general I would recommend to use install Python on your machine and also run the GUI with "python somedo.py".
Examples of commmand line use:
"python somedo.py Facebook -t holly.gennero -l Email=[email protected],[email protected] Password=nakatomi -o Timeline=True"
It is possible to use a file which defines the jobs:
jobfile.txt
Facebook -l Email=[email protected],[email protected] Password=nakatomi -o Photos=True Instagram -t karl,tony,theo -o Media=True
Start the jobs with "python somedo.py -f jobfile.txt".
Here are tho possible options/parameters
Facebook --target, -t string --login, -l Email=string Password=string --options, -o Timeline=bool (True or False) expandTimeline=bool translateTimeline=bool untilTimeline=string (format: Y-m-d, e.g. 2017-12-31) limitTimeline=int (e.g. 50) About=bool Photos=bool expandPhotos=bool translatePhotos=bool limitPhotos=int Network=bool depthNetwork=int extendNetwork=bool
Instagram --options, -o Media=bool limitPages=int
Twitter, --options, -o Search=bool Photos=bool limitPages=int
Debug modes DEBUG and VISIBLE can be set as first command line argument, e.g.:
"python somedo.py VISIBLE" starts the GUI and will show Chrome/Chromium while running.
"python somedo.py DEBUG Twitter -t nakatomi" start a Twitter job in debug mode.
5 Development and License
The use, development, distribution, etc. of the script is subject to the restrictions of GPL Version 3. If you are interested in the code, visit the project on Github:
https://github.com/markusthilo/somedo
The network visualisation uses Vis.js which is dual licensed under both Apache 2.0 and MIT:
http://visjs.org http://www.apache.org/licenses/LICENSE-2.0 http://opensource.org/licenses/MIT
The tool is in alpha state and propably will not be stable enough to have a final release ever. Constant changes and inhomogeneity of pages require permanent adaptation and development. The completeness and correctness of the information obtained can never be guaranteed. The developer is not responsible for the use of the tool. You are welcome to participate or donate to the development. Feel free to report bugs or give suggestions by email to: