Skip to content

feat(probes): add code-switching jailbreak probe (codeswitch) #1995

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Draft
Yugesh-reddy wants to merge 3 commits into
base: main
Choose a base branch
from
Draft

feat(probes): add code-switching jailbreak probe (codeswitch) #1995

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
3 commits
Select commit Hold shift + click to select a range
0b83fca
feat(probes): add code-switching jailbreak probe (codeswitch)
Yugesh-reddy Jul 14, 2026
fb8309e
docs(probes): point codeswitch citation at MediCS project
Yugesh-reddy Jul 28, 2026
6416f50
fix(probes): set codeswitch lang to * to skip translation
Yugesh-reddy Jul 28, 2026
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
1 change: 1 addition & 0 deletions docs/source/index_probes.rst
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -20,6 +20,7 @@ For a guide to writing probes, see :doc:`extending.probe`.
probes/av_spam_scanning
probes/badchars
probes/base
probes/codeswitch
probes/continuation
probes/dan
probes/divergence
Expand Down
9 changes: 9 additions & 0 deletions docs/source/probes/codeswitch.rst
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -0,0 +1,9 @@
garak.probes.codeswitch
=======================

.. automodule:: garak.probes.codeswitch
:members:
:undoc-members:
:show-inheritance:

.. show-asr::
10 changes: 10 additions & 0 deletions garak/data/codeswitch/README.md
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -0,0 +1,10 @@
# Code-switch probe prompts

Prompts in `codeswitch_multi.txt` are the MediCS-500 attack prompts
(intra-sentential medical-domain code-switching across gu/hi/sw/tl/yo/bn).

Source project: https://github.com/Yugesh-reddy/MediCS-Red-Teaming

These rows are author-contributed for use in garak under the same Apache-2.0
terms as the rest of this repository. Do not regenerate or expand this file
with machine translation; keep prompts verbatim.
Loading