Skip to content

Releases: ivan-sincek/forbidden

v12.6

28 Sep 22:10
Compare
Choose a tag to compare

v12.6 release notes:

Breaking changes sincek v12.3!

  • removed redundant dependency in the TOML file,
  • version bump.

Python tool for brute forcing 4xx response status codes. Based on PycURL and Python Requests.

v12.5

20 Sep 16:39
Compare
Choose a tag to compare

v12.5 release notes:

Breaking changes sincek v12.3!

  • removed unused imports,
  • version bump.

Python tool for brute forcing 4xx response status codes. Based on PycURL and Python Requests.

v12.4

15 Sep 16:15
Compare
Choose a tag to compare

v12.4 release notes:

Breaking changes sincek v12.3!

  • added missing time import for -s / --sleep option in forbidden,
  • other small bug fixes.

Python tool for brute forcing 4xx response status codes. Based on PycURL and Python Requests.

v12.3

12 Sep 18:16
Compare
Choose a tag to compare

v12.3 release notes:

Breaking changes!

  • new test named values - where you can test HTTP request headers using only your own values,
  • -v / --values option now also accepts a single value for quick testing - not only a single file,
  • paths-sniper test is renamed to paths-ram - breaking change,
  • new option -dmp / --dump - where you can just dump all the test cases in the output file without actually running them,
  • new option -sc / --status-codes - where you can specify which HTTP response status codes to include in the results,
  • by default, stresser is now "locked" to Python Requests instead of PycURL, while forbbiden stays the same, "locked" to PycURL,
  • in streser, -ic / --ignore-curl option is renamed to -ir / --ignore-requests - breaking change,
  • added error handling for HTTP response decoding, and also added OSError in the existing exception handlers - mostly because of PycURL complaining,
  • added a new fancy progress bar,
  • refactored some code.

Python tool for brute forcing 4xx response status codes. Based on PycURL and Python Requests.

v11.2

15 May 11:30
Compare
Choose a tag to compare

v11.2 release notes:

  • fixed a bug when generating paths-sniper tests when there is no directory in the URL,
  • improved methods for sending requests by adding more control over what is returned,
  • optimized the code,
  • version bump.

Python tool for brute forcing 4xx response status codes. Based on PycURL and Python Requests.

v11.0

13 May 10:02
Compare
Choose a tag to compare

v11.0 release notes:

  • fixed a bug where testing double headers was not properly locked to Python Requests,
  • added new options allowing you to set multiple additional HTTP request headers using -H and HTTP cookies using -b,
  • added new path bypasses and a new and lighter paths-sniper test case,
  • added a table output - nice for analyzing results on wide screens,
  • improved and optimized the code,
  • version bump.

Python tool for brute forcing 4xx response status codes. Based on PycURL and Python Requests.

v10.9

11 May 10:24
Compare
Choose a tag to compare

v10.9 release notes:

  • small text changes in README,
  • removed redundant argparse dependency (thanks for pointing it out fabaff and branchvincent),
  • updated the user agents list,
  • version bump.

Python tool for brute forcing 4xx response status codes. Based on PycURL and Python Requests.

v10.8

08 Mar 14:21
Compare
Choose a tag to compare

v10.8 release notes:

  • install instructions update,
  • updated the user agents list,
  • added AWS IP 169.254.169.254,
  • version bump.

Python tool for brute forcing 4xx response status codes. Based on PycURL and Python Requests.

v10.6

12 Jan 15:28
Compare
Choose a tag to compare

v10.6 release notes:

  • added request timeout (-rt) option.

Python tool for brute forcing 4xx response status codes. Based on PycURL and Python Requests.

v10.5

29 Dec 21:34
Compare
Choose a tag to compare

v10.5 release notes:

  • argument parsing code rebase and improved few other stuff,
  • boolean options / arguments no longer require yes or no value.

Python tool for brute forcing 4xx response status codes. Based on PycURL and Python Requests.