pip3 install passphraseme
passphraseme with a number to generate secure passphrases using EFF's
large wordlist, like this:
$ passphraseme 7 banana stopwatch appealing germinate survival retired comma $ passphraseme 5 borrower harvest stature entity blimp
You can also optionally choose a different wordlist. Here are all of the command line arguments:
||show help message|
||Separator (default " ")|
||Use EFF's general short wordlist|
||Use EFF's short wordlist with unique prefixes|
||Use EFF's Game of Thrones wordlist (Passwords of Westeros)|
||Use EFF's Harry Potter wordlist (Accio Passphrase!)|
||Use EFF's Star Trek wordlist (Live Long and Passphrase)|
||Use EFF's Star Wars wordlist (The Passphrase Is Strong With This One)|
||Custom wordlist filename|
For example, you can choose to use one of EFF's short wordlists, like this:
$ passphraseme -s1 5 glide canal flag sage those $ passphraseme -s2 5 optical anonymous nirvana agitate feudalist
Or you can embrace your inner nerd and use a fandom wordlist:
$ passphraseme --game-of-thrones 5 crow betrayed severely gloating asked $ passphraseme --harry-potter 5 mirror relief date future mysterious $ passphraseme --star-trek 5 children refused captain cornwell vulcan $ passphraseme --star-wars 5 unkar struggle names ally cantina
You can also choose to use a custom wordlist, like this:
$ passphraseme -d /usr/share/dict/words 7 leading's Oz's caesareans lactate eloped interposed wowed
And if you prefer, you can use a custom separator, like
- instead of
$ passphraseme --sep - 5 ungreased-tried-broadcast-deduce-yield $ passphraseme --sep . 5 parmesan.unkempt.budget.premiere.puritan
Strength of passphrases
This table shows the strength (bits of entropy) of
passphraseme-generated passphrases of different lengths (1-10 words).
|Bits of entropy/word||1||2||3||4||5||6||7||8||9||10|
|EFF large wordlist (default)||12.925||12.9 (0 s)||25.8 (0 s)||38.8 (0 s)||51.7 (1 h)||64.6 (1 y)||77.5 (10.6k y)||90.5 (82M y)||103.4 (642B y)||116.3 (4.99e15 y)||129.2 (3.88e19 y)|
|EFF short wordlists||10.339||10.3 (0 s)||20.7 (0 s)||31.0 (0 s)||41.4 (4 s)||51.7 (1 h)||62.0 (83 d)||72.4 (295 y)||82.7 (382.3k y)||93.1 (495M y)||103.4 (642B y)|
|EFF fandom wordlists||11.965||12.0 (0 s)||23.9 (0 s)||35.9 (0 s)||47.9 (6 m)||59.8 (17 d)||71.8 (196 y)||83.8 (787.1k y)||95.7 (3B y)||107.7 (1.26e13 y)||119.7 (5.04e16 y)|
The brute force time is calculated like this:
I'm assuming you're using a passphrase for macOS 10.8+ (PBKDF2-SHA512) to encrypt your disk with FileVault. According to this post, the password cracking tool hashcat can guess 193,900 passphrases per second on an Amazon AWS p3.16xlarge instance, which costs $24.48 per hour.
If an attacker is willing to spend up to $1 billion per day to guess your passphrase, they can afford to run 1.7 million of these AWS instances at once, meaning they can guess ~330 billion passphrases per second. On average, a brute force attack will find the passphrase after searching half the keyspace, so the times above are how long it takes to search half the keyspace.
Note that the time "3.88e19 y" means "3.88 x 1019 years". Also note that the brute force times will vary wildly, both much quicker or much slower, depending on the hash function or KDF used -- basically, depending on what software you're using this passphrase with.
Check out calc_passphrase_strength.py to see the maths.
The wordlists included were created by Electronic Frontier Foundation, and are distributed under the Creative Commons Attribution 3.0. For the fandom wordlists (Game of Thrones, Harry Potter, Star Trek, and Star Wars), EFF notes that "Any trademarks within the word list are the property of their respective trademark holders, who are not affiliated with the Electronic Frontier Foundation and do not sponsor or endorse these passwords."