Official datasetProvider-published evidence
WorkflowEvals Security Incidents
A 240-case dataset for security-alert triage, containment, and action selection.
The official dataset organizes security incident cases, policy questions, and published decisions for reproducible workflow-level evaluation.
- Publisher
- TypeSafe AI
- Published or submitted
- Not specified by the source
- Last verified
- October 8, 2026
Type Safe catalogs sources and evidence boundaries. Provider datasets and third-party preprints do not become an independent leaderboard.
Key facts
- Cases
- 240
- Question instances
- 1,820
- Published run results
- 2,160
- Dataset license
- Not specified in the reviewed source
Task and scope
Security-alert triage and containment.
Dataset schema
cases · questions · run_results · 4,220 rows
Evaluation metrics
exact_actions · primary_action
Dataset license
Not specified by the source
Method and usage
- The workflow evaluates security-alert triage and containment decisions against selected reference actions.
- Exact action sets include arguments; primary_action compares the main selected action where available.
Limitations and evidence boundaries
- The reviewed dataset card did not specify a dataset license.
- Synthetic or curated security cases cannot substitute for organization-specific threat modeling and human incident response.