FEAT: add benchmark-owned adversarial prompts - #2494
Open
hannahwestra25 wants to merge 16 commits into
Open
Conversation
Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com> Copilot-Session: 545c2d8d-1fe9-451d-a8bc-da23e19b3737
Replace the user-supplied prompt override with scenario-local prompt variants for the benchmark's role-play, simulated Crescendo, and TAP techniques. Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com> Copilot-Session: 545c2d8d-1fe9-451d-a8bc-da23e19b3737
hannahwestra25
force-pushed
the
hannahwestra25-scenario-adversarial-system-prompt
branch
from
August 25, 2026 21:09
62452c8 to
84be7c0
Compare
hannahwestra25
changed the title
Adapt the supplied single-turn and multi-turn prompts to the adversarial_chat schema and prepend them to the corresponding benchmark technique prompts. Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com> Copilot-Session: 545c2d8d-1fe9-451d-a8bc-da23e19b3737
Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com> Copilot-Session: 545c2d8d-1fe9-451d-a8bc-da23e19b3737
hannahwestra25
changed the title
This reverts commit d4b29b5.
hannahwestra25
changed the title
Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com> Copilot-Session: 545c2d8d-1fe9-451d-a8bc-da23e19b3737
This reverts commit 0f96620.
hannahwestra25
changed the title
hannahwestra25
changed the title
hannahwestra25
changed the title
hannahwestra25
marked this pull request as ready for review
August 26, 2026 16:39
hannahwestra25
commented
Aug 26, 2026
| @@ -0,0 +1,84 @@ | |||
| name: adversarial_benchmark_crescendo_simulated | |||
| description: | | |||
Merge current origin/main and replace duplicated benchmark prompts with a shared guidance layer composed over canonical technique prompts. Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com>
Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com>
Apply static benchmark guidance to every selected adversarial factory through scenario-local copies while preserving native prompts and legacy path-backed seeds. Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com> Copilot-Session: 86af64f1-e02f-45f9-9cc8-15325ab1644e
…5-scenario-adversarial-system-prompt
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
What changed
AdversarialBenchmarknow prepends one benchmark-owned, cross-technique guidance layer to the native adversarial system prompt of every selected compatible attack technique, including explicitly selected techniques such asred_teaming.The guidance is applied through scenario-local factory copies, so globally registered factories and canonical technique prompts remain unchanged. Static layers preserve the native prompt's template parameters, response schema, data type, and response format, and participate in attack/technique identity.
Simulated-conversation techniques retain their existing path-backed prompt configuration. Prefixes are stored in the existing JSON-backed seed value, so legacy records remain compatible and no database migration is required.
The default techniques remain
role_play_video_game,crescendo_simulated, andtap. The scenario version is bumped to 5 because adversarial guidance changes benchmark behavior.Testing