Hackathon demo: adversarial benchmark pipeline + metrics dashboard - #2684
Merged
hannahwestra25 merged 1 commit intoSep 16, 2026
Conversation
…to adversarial dataset branch Transplants the dashboard-only work from PR microsoft#2667 (hannahwestra25-metrics-dashboard) onto this branch's already-integrated benchmark-store/cache-reuse/pipeline-automation work from PR microsoft#2551, producing a single hackathon-demo branch with the full data -> pipeline -> committed store -> dashboard loop. Brings over, unchanged from microsoft#2667: - doc/dashboard/ (0_dashboard.md, 1_scorer_quality.md, 2_benchmark_leaderboard.md, and the committed _generated/*.html fragments) - build_scripts/generate_dashboard_html.py (renders the committed JSONL/scorer_evals data into the static HTML fragments the docs pages include) - build_scripts/import_adversarial_benchmark_snapshot.py + its test (manual fallback to upsert a technique-metrics.json snapshot, e.g. downloaded from an ADO pipeline artifact, when the scenario result isn't in local memory) - pyrit/datasets/benchmark_results/adversarial_benchmark_metrics.jsonl seeded with real ADO-ingested demo data - doc/myst.yml (nav entry) and pyrit/common/path.py (BENCHMARK_RESULTS_PATH constant) Deliberately NOT brought over: the exporter identity-fields/cache-reuse logic and scenario/pipeline changes from microsoft#2667's f973f5e, since this branch already has its own, more advanced version of that same logic (55bece4, cherry-picked earlier from f973f5e and reconciled against this branch's dynamic BenchmarkTechnique registry). Re-applying microsoft#2667's version here would conflict with and regress that integration. Also updates 0_dashboard.md's "Refreshing the data" section: it previously said there was no CI automation for the benchmark store; this branch's pipeline now has an opt-in publishBenchmarkStore parameter that runs the exporter and opens a PR automatically, so the doc now describes that path alongside the fully-manual one. Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com>
hannahwestra25
marked this pull request as ready for review
September 16, 2026 20:00
hannahwestra25
merged commit Sep 16, 2026
e8f64ff
into
microsoft:copilot/adversarial_dataset
1 check passed
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.
Combines PR #2551's benchmark-store/cache-reuse/pipeline-automation work with PR #2667's metrics dashboard (scorer quality + benchmark leaderboard) into one branch for an end-to-end hackathon demo. Not intended to merge as-is — opened so the
.azuredevops/adversarial-benchmark.ymlpipeline can be manually triggered against a realmicrosoft/PyRITbranch.