Skip to content

Add special chars injection detector (WIP) #1309

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

Closed
wants to merge 137 commits into from
Closed

Add special chars injection detector (WIP) #1309

Show file tree
Hide file tree
Changes from 1 commit
Commits
Show all changes
137 commits
Select commit Hold shift + click to select a range
5bb1dbb
Add special chars injection detector (WIP)
mattbit Aug 9, 2023
85de0d5
Merge branch 'main' into scan-special-chars-injection
mattbit Aug 10, 2023
69345ac
Merge branch 'main' into scan-special-chars-injection
mattbit Aug 18, 2023
5c8c30c
Update and register control chars detector
mattbit Aug 18, 2023
339b685
Merge branch 'main' into scan-special-chars-injection
mattbit Aug 28, 2023
0d545eb
WIP: readme to wrap a custom generative model
kevinmessiaen Aug 30, 2023
8ea0914
Added model description
kevinmessiaen Aug 30, 2023
1c5f8f2
Added "business" detector
kevinmessiaen Aug 30, 2023
cd24db9
Merge branch 'main' into poc/scan_llm
kevinmessiaen Aug 30, 2023
9da77e6
WIP: llm documentation querying notebook
kevinmessiaen Aug 31, 2023
784eb35
Updated notebook
kevinmessiaen Aug 31, 2023
aa1b601
Deleted test notebook
kevinmessiaen Aug 31, 2023
9cab5de
Merge branch 'main' into scan-special-chars-injection
mattbit Aug 31, 2023
3842090
Fixing test and refactoring
kevinmessiaen Sep 1, 2023
4759345
Added method to get the generated dataset
kevinmessiaen Sep 1, 2023
02a47b6
Fixed NaN in inferred dataset
kevinmessiaen Sep 1, 2023
ab614de
Typo
kevinmessiaen Sep 1, 2023
d7e9df8
Merge branch 'main' into poc/scan_llm
kevinmessiaen Sep 1, 2023
aa4362b
Updated example notebook
kevinmessiaen Sep 1, 2023
752f104
Added notebook in text-generation and text guides
kevinmessiaen Sep 1, 2023
0807cad
Added dist
kevinmessiaen Sep 1, 2023
1936d56
Merge branch 'main' into poc/scan_llm
kevinmessiaen Sep 1, 2023
a1b1d67
Updated wheel
kevinmessiaen Sep 1, 2023
e3aea1f
Added missing clean dist in gradle config
kevinmessiaen Sep 4, 2023
094ecbd
Added LLM issue categories
kevinmessiaen Sep 4, 2023
07cd18f
Merge branch 'main' into poc/scan_llm
kevinmessiaen Sep 4, 2023
b2e6d66
Removed unused business detector
kevinmessiaen Sep 4, 2023
95a85ad
Refactored business detector issues
kevinmessiaen Sep 4, 2023
7643a29
Merge branch 'poc/talk-to-my-ml' into poc/scan_llm
kevinmessiaen Sep 4, 2023
8aa952c
Use configured llm for scan
kevinmessiaen Sep 4, 2023
ea8ef3d
Use configured llm for scan
kevinmessiaen Sep 4, 2023
784f0e0
WIP: Reformatting scanner
kevinmessiaen Sep 4, 2023
bff8b6b
Fixed issue
kevinmessiaen Sep 4, 2023
3f13230
Fixed issue
kevinmessiaen Sep 4, 2023
b957ba7
Fixed typo
kevinmessiaen Sep 4, 2023
b401326
Merge branch 'main' into poc/scan_llm
kevinmessiaen Sep 5, 2023
be62fe7
Removed the explanation feature for text generative application
kevinmessiaen Sep 5, 2023
b1894f3
Fixed code smell
kevinmessiaen Sep 5, 2023
e726666
Code improvement
kevinmessiaen Sep 5, 2023
c6b955e
Fixed import
kevinmessiaen Sep 5, 2023
0531065
Fixed worker for text generation
kevinmessiaen Sep 5, 2023
9c53f2c
Display generative results
kevinmessiaen Sep 6, 2023
7c4c12d
Code cleanup
kevinmessiaen Sep 6, 2023
9866663
Add generate button to regenerate answer
kevinmessiaen Sep 6, 2023
3620b78
Fixed display of newlines
kevinmessiaen Sep 6, 2023
7de4e4e
WIP
kevinmessiaen Sep 7, 2023
6ce31c6
Rename GENERATE to RUN MODEL in LLM debugger UI
kevinmessiaen Sep 7, 2023
3c8b2b2
WIP: added categories
kevinmessiaen Sep 7, 2023
02fe9ae
Fixed scan prompt issue
kevinmessiaen Sep 7, 2023
af67628
Changed default model to improve robustness
kevinmessiaen Sep 7, 2023
21cd1eb
Limit test case per issue
kevinmessiaen Sep 8, 2023
06b6875
Fixed desc of Issue
kevinmessiaen Sep 8, 2023
0c8f10a
Fixed test
kevinmessiaen Sep 8, 2023
4a90228
Removed old detectors from scan
kevinmessiaen Sep 11, 2023
e8e2420
Improved scan robustness with more customized prompts
kevinmessiaen Sep 11, 2023
6b4f796
Fixed import
kevinmessiaen Sep 11, 2023
0759845
Improved test description
kevinmessiaen Sep 12, 2023
dad9836
Fixed minor/major
kevinmessiaen Sep 12, 2023
cb14160
Improved test generation
kevinmessiaen Sep 12, 2023
a956a41
Prompt engineering
kevinmessiaen Sep 13, 2023
a86fad8
Merge branch 'poc/talk-to-my-ml' into poc/scan_llm
kevinmessiaen Sep 13, 2023
fadbf53
Added explanation and tips + clearer desc for LLM auto scan
kevinmessiaen Sep 13, 2023
2af8033
Added reason and tip columns
kevinmessiaen Sep 13, 2023
90f2a41
Updated notebook
kevinmessiaen Sep 14, 2023
235e163
Merge branch 'poc/talk-to-my-ml' into poc/scan_llm
kevinmessiaen Sep 14, 2023
3ea9961
Fixed tests
kevinmessiaen Sep 14, 2023
dfd4ed0
Merge branch 'poc/talk-to-my-ml' into poc/scan_llm
kevinmessiaen Sep 14, 2023
ea49e8c
Merge branch 'main' into poc/scan_llm
kevinmessiaen Sep 14, 2023
2da6f1f
Fixing tests
kevinmessiaen Sep 14, 2023
0c46949
Merge branch 'main' into poc/scan_llm
andreybavt Sep 14, 2023
7911c40
Fixed test_default_dataset_is_used_with_generative_model
kevinmessiaen Sep 15, 2023
00f1021
Merge branch 'main' into poc/scan_llm
kevinmessiaen Sep 15, 2023
289adf8
Updated article summary notebook and improved support of langchain LL…
kevinmessiaen Sep 15, 2023
2eeac46
Fixed issue
kevinmessiaen Sep 15, 2023
18a914f
Updated notebook
kevinmessiaen Sep 15, 2023
8fd53b3
Merge branch 'main' into poc/scan_llm
kevinmessiaen Sep 18, 2023
efe2a27
Merge remote-tracking branch 'origin/poc/scan_llm' into poc/scan_llm
kevinmessiaen Sep 18, 2023
ef918e1
Skip LLM test
kevinmessiaen Sep 18, 2023
076a425
Merge branch 'main' into poc/scan_llm
kevinmessiaen Sep 18, 2023
198e19b
Improved result of scan
kevinmessiaen Sep 18, 2023
f281728
Merge branch 'main' into poc/scan_llm
kevinmessiaen Sep 19, 2023
8d1a023
Improved result of scan and example with how to reuse suite
kevinmessiaen Sep 19, 2023
5a4ed34
Merge remote-tracking branch 'origin/poc/scan_llm' into poc/scan_llm
kevinmessiaen Sep 19, 2023
f0ddc35
Code cleanup and usage of maybe_print
kevinmessiaen Sep 19, 2023
29dbe91
Scan prompt optimization
kevinmessiaen Sep 19, 2023
401d935
Merge branch 'main' into poc/scan_llm
andreybavt Sep 19, 2023
fdb7388
Merge branch 'main' into poc/scan_llm
mattbit Sep 20, 2023
4c0e15b
Fixed generation of inputs
kevinmessiaen Sep 22, 2023
0fc567c
Fixed generation of inputs
kevinmessiaen Sep 22, 2023
7310d7c
Usage of openai functions for output formatting
kevinmessiaen Sep 22, 2023
20a784f
Improvements
kevinmessiaen Sep 22, 2023
479c064
Better issue displaying
kevinmessiaen Sep 22, 2023
81f8ead
Improved scan with better input generation
kevinmessiaen Sep 22, 2023
e449568
Use GPT4 for validation
kevinmessiaen Sep 22, 2023
995cdbb
Added format instruction and examples
kevinmessiaen Sep 22, 2023
059ea91
Improved generate_dataset
kevinmessiaen Sep 25, 2023
21dee4e
Merge branch 'main' into poc/scan_llm
kevinmessiaen Sep 25, 2023
3176a5a
fix merge
kevinmessiaen Sep 25, 2023
1730971
Updated hallucination examples
kevinmessiaen Sep 25, 2023
d731b4a
Updated hallucination examples
kevinmessiaen Sep 25, 2023
b67c5fa
Grammar
kevinmessiaen Sep 25, 2023
90bab22
Removed validation
kevinmessiaen Sep 25, 2023
074cd81
Default GPT 3
kevinmessiaen Sep 25, 2023
2c17098
Fixed metric of tests
kevinmessiaen Sep 25, 2023
af11cff
WIP: LLM Hub doc
kevinmessiaen Sep 26, 2023
8747b97
Merge branch 'main' into poc/scan_llm
kevinmessiaen Sep 27, 2023
68cc580
Generate dataset name and renamed Test case to LLM response validati…
kevinmessiaen Sep 27, 2023
3cb14af
Add display name of tests
kevinmessiaen Sep 27, 2023
7f79332
Improved argument extractor
kevinmessiaen Sep 27, 2023
aaa3ac4
Improved docstring
kevinmessiaen Sep 27, 2023
b9bf09a
Use text area for string input to improve UX
kevinmessiaen Sep 27, 2023
5aae7f2
Added single evaluation test
kevinmessiaen Sep 27, 2023
7c8a7a9
Merge remote-tracking branch 'origin/main' into poc/scan_llm
kevinmessiaen Sep 28, 2023
4400cd1
Automatically import scan detectors
mattbit Sep 29, 2023
c95d0ec
Split LLM scan detectors
mattbit Sep 29, 2023
7f189c6
Prototype of sycophancy detection
mattbit Oct 2, 2023
98c022b
Scan template style adjustments
mattbit Oct 2, 2023
2297c72
Add implausible output check in hallucination detection
mattbit Oct 2, 2023
a64ed68
v0 of garak integration
rabah-khalek Oct 3, 2023
3b747d4
updated resources init
rabah-khalek Oct 3, 2023
63f497c
updated resources init
rabah-khalek Oct 3, 2023
a32815c
added DAN to prompt injection detector
rabah-khalek Oct 4, 2023
7f5cdc0
failure rate corrected
rabah-khalek Oct 4, 2023
0465d0b
updating detector
rabah-khalek Oct 4, 2023
87d139a
Refine hallucination detectors
mattbit Oct 4, 2023
38ebbe7
Update scan widget styles
mattbit Oct 5, 2023
f9c77aa
Merge branch 'feature/llm-scan' into GSK-1816-prompt-injection-garak
rabah-khalek Oct 5, 2023
0dde501
added automatic inference of input_variable
rabah-khalek Oct 5, 2023
48bc036
polishing
rabah-khalek Oct 5, 2023
b7f0c17
added colorama, 🔒 pdm Lock updated
rabah-khalek Oct 5, 2023
e6b9157
Merge branch 'feature/llm-scan' into GSK-1816-prompt-injection-garak
rabah-khalek Oct 6, 2023
dc2d265
Prompt engineering and more function calling
mattbit Oct 6, 2023
dd68c75
Merge branch 'feature/llm-scan' into GSK-1816-prompt-injection-garak
mattbit Oct 6, 2023
8e33929
Add LICENSE
mattbit Oct 6, 2023
af6c009
Merge pull request #1460 from Giskard-AI/GSK-1816-prompt-injection-garak
mattbit Oct 6, 2023
36ee5bd
Merge branch 'feature/llm-scan' into scan-special-chars-injection
mattbit Oct 6, 2023
6dfbfaa
Fix merge
mattbit Oct 6, 2023
File filter

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
Prev Previous commit
Next Next commit
added colorama, 🔒 pdm Lock updated
  • Loading branch information
@rabah-khalek
rabah-khalek committed Oct 5, 2023
commit b7f0c17e86f29d3ba5aa647048861f696b31b60f
Loading