FEAT Add HarmfulQA dataset loader by romanlutz · Pull Request #1421 · microsoft/PyRIT

romanlutz · 2026-03-01T14:14:38Z

Add remote dataset loader for HarmfulQA (declare-lab/HarmfulQA), containing ~2k harmful questions organized by academic topic and subtopic for testing LLM susceptibility to harm-inducing question-answering.

Copilot

Pull request overview

Adds a new remote seed-dataset provider for the HuggingFace declare-lab/HarmfulQA dataset so it can be fetched and used as SeedPrompt entries within PyRIT’s dataset discovery/registration system.

Changes:

Introduced _HarmfulQADataset remote loader that fetches HarmfulQA from HuggingFace and converts rows into SeedPrompts.
Exported the new loader from pyrit.datasets.seed_datasets.remote to trigger auto-registration.
Added unit tests validating basic fetch + conversion behavior and dataset_name.

Reviewed changes

Copilot reviewed 4 out of 4 changed files in this pull request and generated 2 comments.

File	Description
pyrit/datasets/seed_datasets/remote/harmful_qa_dataset.py	New remote dataset loader implementation for HarmfulQA -> `SeedDataset`/`SeedPrompt` conversion.
pyrit/datasets/seed_datasets/remote/init.py	Re-export/import the new loader so it’s discoverable/registered alongside other remote loaders.
tests/unit/datasets/test_harmful_qa_dataset.py	Unit tests for fetching/conversion and `dataset_name` behavior.

pyrit/datasets/seed_datasets/remote/harmful_qa_dataset.py

Add remote dataset loader for HarmfulQA (declare-lab/HarmfulQA), containing ~2k harmful questions organized by academic topic and subtopic for testing LLM susceptibility to harm-inducing question-answering. Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>

Copilot

Pull request overview

Copilot reviewed 4 out of 4 changed files in this pull request and generated 3 comments.

doc/code/datasets/1_loading_datasets.ipynb

The HF dataset identifier is now a class constant HF_DATASET_NAME instead of a constructor parameter, consistent with other loaders. Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>

Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>

Copilot

Pull request overview

Copilot reviewed 4 out of 4 changed files in this pull request and generated 2 comments.

tests/unit/datasets/test_harmful_qa_dataset.py

doc/code/datasets/1_loading_datasets.ipynb

Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>

Copilot

Copilot encountered an error and was unable to review this pull request. You can try again by re-requesting a review.

…-qa-dataset # Conflicts: # doc/code/datasets/1_loading_datasets.ipynb

Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>

Copilot

Pull request overview

Copilot reviewed 4 out of 4 changed files in this pull request and generated 1 comment.

pyproject.toml

Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>

…-qa-dataset # Conflicts: # pyproject.toml

Copilot

Pull request overview

Copilot reviewed 3 out of 3 changed files in this pull request and generated no new comments.

…-qa-dataset

Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>

Copilot

Pull request overview

Copilot reviewed 4 out of 4 changed files in this pull request and generated no new comments.

…-qa-dataset

varunj-msft · 2026-03-03T18:18:16Z

Reviewed the PR - code follows established patterns, tests pass,, fetches all seeds from HuggingFace successfully. Looks good to me, ready to merge.

…-qa-dataset

Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>

Copilot

Pull request overview

Copilot reviewed 5 out of 5 changed files in this pull request and generated no new comments.

You can also share your feedback on Copilot code review. Take the survey.

…-qa-dataset # Conflicts: # doc/code/datasets/1_loading_datasets.ipynb

Includes beaver_tails and harmful_qa in the dataset listing. Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>

Copilot AI review requested due to automatic review settings March 1, 2026 14:14

Copilot started reviewing on behalf of romanlutz March 1, 2026 14:15 View session

romanlutz force-pushed the romanlutz/add-harmful-qa-dataset branch from f8de803 to e996238 Compare March 1, 2026 14:16

Copilot AI reviewed Mar 1, 2026

View reviewed changes

pyrit/datasets/seed_datasets/remote/harmful_qa_dataset.py Show resolved Hide resolved

pyrit/datasets/seed_datasets/remote/harmful_qa_dataset.py Outdated Show resolved Hide resolved

romanlutz force-pushed the romanlutz/add-harmful-qa-dataset branch from e996238 to d441180 Compare March 1, 2026 14:26

Copilot AI review requested due to automatic review settings March 2, 2026 13:00

romanlutz force-pushed the romanlutz/add-harmful-qa-dataset branch from d441180 to b4c033f Compare March 2, 2026 13:00

Copilot started reviewing on behalf of romanlutz March 2, 2026 13:01 View session

Copilot AI reviewed Mar 2, 2026

View reviewed changes

doc/code/datasets/1_loading_datasets.ipynb Outdated Show resolved Hide resolved

doc/code/datasets/1_loading_datasets.ipynb Show resolved Hide resolved

doc/code/datasets/1_loading_datasets.ipynb Show resolved Hide resolved

romanlutz and others added 2 commits March 2, 2026 05:36

Remove dataset_name from constructor, hardcode as class constant

bf04ed3

The HF dataset identifier is now a class constant HF_DATASET_NAME instead of a constructor parameter, consistent with other loaders. Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>

Make authors multi-line, guard empty harm_categories

a2ad023

Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>

Copilot AI review requested due to automatic review settings March 2, 2026 13:46

Copilot started reviewing on behalf of romanlutz March 2, 2026 13:46 View session

Copilot AI reviewed Mar 2, 2026

View reviewed changes

tests/unit/datasets/test_harmful_qa_dataset.py Outdated Show resolved Hide resolved

doc/code/datasets/1_loading_datasets.ipynb Outdated Show resolved Hide resolved

romanlutz and others added 2 commits March 2, 2026 05:53

Use AsyncMock for _fetch_from_huggingface in tests

03156fa

Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>

Precompute source_url and groups outside the loop

0a75a74

Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>

Copilot AI review requested due to automatic review settings March 2, 2026 14:02

romanlutz and others added 2 commits March 2, 2026 06:05

Wrap prompt values in raw/endraw to preserve Jinja2 syntax

4989e53

Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>

Add license notice and content warning to docstring

e3998d5

Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>

Copilot AI reviewed Mar 2, 2026

View reviewed changes

romanlutz and others added 3 commits March 2, 2026 14:44

Merge remote-tracking branch 'origin/main' into romanlutz/add-harmful…

4beab37

…-qa-dataset # Conflicts: # doc/code/datasets/1_loading_datasets.ipynb

fix: update notebook output with all dataset names

90d6796

add E402/E501 to doc per-file-ignores

24758ab

Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>

Copilot AI review requested due to automatic review settings March 3, 2026 00:50

Copilot started reviewing on behalf of romanlutz March 3, 2026 00:51 View session

Copilot AI reviewed Mar 3, 2026

View reviewed changes

pyproject.toml Outdated Show resolved Hide resolved

romanlutz and others added 2 commits March 2, 2026 19:56

fix: update per-file-ignores comment to reflect E402/E501

2a86621

Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>

Merge remote-tracking branch 'origin/main' into romanlutz/add-harmful…

f11b0a1

…-qa-dataset # Conflicts: # pyproject.toml

Copilot AI review requested due to automatic review settings March 3, 2026 04:50

Copilot started reviewing on behalf of romanlutz March 3, 2026 04:50 View session

Copilot AI reviewed Mar 3, 2026

View reviewed changes

romanlutz and others added 2 commits March 2, 2026 21:01

Merge remote-tracking branch 'origin/main' into romanlutz/add-harmful…

c4efaef

…-qa-dataset

fix: update per-file-ignores comment to match actual rules

1dde407

Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>

Copilot AI review requested due to automatic review settings March 3, 2026 05:04

Copilot started reviewing on behalf of romanlutz March 3, 2026 05:04 View session

Copilot AI reviewed Mar 3, 2026

View reviewed changes

Merge remote-tracking branch 'origin/main' into romanlutz/add-harmful…

8925a8f

…-qa-dataset

Merge remote-tracking branch 'origin/main' into romanlutz/add-harmful…

8246f5f

…-qa-dataset

Copilot AI review requested due to automatic review settings March 4, 2026 06:03

Copilot started reviewing on behalf of romanlutz March 4, 2026 06:03 View session

Regenerate 1_loading_datasets.ipynb via jupytext

c95ea32

Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>

Copilot AI reviewed Mar 4, 2026

View reviewed changes

hannahwestra25 approved these changes Mar 4, 2026

View reviewed changes

hannahwestra25 self-assigned this Mar 4, 2026

romanlutz and others added 2 commits March 4, 2026 13:03

Merge remote-tracking branch 'origin/main' into romanlutz/add-harmful…

c483d92

…-qa-dataset # Conflicts: # doc/code/datasets/1_loading_datasets.ipynb

Regenerate loading datasets notebook showing all datasets

b0b1824

Includes beaver_tails and harmful_qa in the dataset listing. Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>

jbolor21 approved these changes Mar 5, 2026

View reviewed changes

romanlutz merged commit e3f03c2 into microsoft:main Mar 5, 2026
38 checks passed

riyosha pushed a commit to riyosha/PyRIT that referenced this pull request Mar 24, 2026

FEAT Add HarmfulQA dataset loader (microsoft#1421)

7a0e340

Conversation

romanlutz commented Mar 1, 2026

Uh oh!

Copilot AI left a comment

Choose a reason for hiding this comment

Pull request overview

Reviewed changes

Uh oh!

Uh oh!

Uh oh!

Copilot AI left a comment

Choose a reason for hiding this comment

Pull request overview

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Copilot AI left a comment

Choose a reason for hiding this comment

Pull request overview

Uh oh!

Uh oh!

Uh oh!

Copilot AI left a comment

Choose a reason for hiding this comment

Uh oh!

Copilot AI left a comment

Choose a reason for hiding this comment

Pull request overview

Uh oh!

Uh oh!

Copilot AI left a comment

Choose a reason for hiding this comment

Pull request overview

Uh oh!

Copilot AI left a comment

Choose a reason for hiding this comment

Pull request overview

Uh oh!

varunj-msft commented Mar 3, 2026

Uh oh!

Copilot AI left a comment

Choose a reason for hiding this comment

Pull request overview

Uh oh!

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

5 participants