Releases: OpenSecOps-Org/SOAR
Releases · OpenSecOps-Org/SOAR
v2.4.21
- Fixed bug: determine_type incorrectly suppressed incidents carrying Compliance metadata without SecurityControlId, causing reassigned findings from the Account Reassignment Preprocessor to silently disappear
- Added test coverage for determine_type function (33 tests)
- Added test coverage for pandas dependencies validating CSV conversion and Decimal handling
- Pinned pandas to 2.2.2 to avoid compatibility issues with pandas 2.3.x in Lambda environment
Full Changelog: v2.4.18...v2.4.21
v2.4.18
- Fixed ticket closure for AWS Security Hub behavior change (July 3, 2025)
- AWS changed from ARCHIVING fixed findings to keeping them ACTIVE with NOTIFIED+PASSED status
Full Changelog: v2.4.17...v2.4.18
v2.4.0
- Automatic reassignment of delegated findings to the originating accounts.
Full Changelog: v2.3.5...v2.4.0
v2.3.2
- Enhanced OpenSecOps Execution Role context in AI prompts to recognize reconnaissance activities as unlikely
- Removed code snippet recommendations for incidents involving AWSControlTowerExecution role
- Improved email formatting
- Added CloudWatch alarm context enrichment for Step Functions and Lambda incidents
- Enhanced AI incident analysis with enriched CloudWatch alarm data and execution context
- Improved timestamp parsing accuracy for CloudWatch alarm processing
- Enhanced AI operation error handling: changed States.Timeout to States.ALL for improved Bedrock timeout resilience
Full Changelog: v2.3.1...v2.3.2
v2.2.1
- Added comprehensive state machine error handling across all 30+ autoremediation functions
- Simplified error handling in IAM.8, S3.3, and EC2.6 autoremediations
- Added centralized SetAutoremediationNotDone state for 100% fallback-to-ticketing coverage
- Improved AI prompt clarity for SOAR self-monitoring infrastructure incidents
- Updated incident_infra.txt and weekly_ai_report_0_common.txt to distinguish individual failures from systemic issues
Full Changelog: v2.2.0...v2.2.1
v2.2.0
- AWS Health Reclassifier: Implemented automatic reclassification of AWS Health informational notifications from HIGH/MEDIUM/CRITICAL to INFORMATIONAL severity to prevent false positives in SOAR processing. The new parameter
ReclassifyAWSHealthIncidentswith a default value ofNocontrols this behaviour. - Default lambda runtime memory increased to 512 MB.
Full Changelog: v2.1.3...v2.2.0
v2.1.3
- Comprehensive testing infrastructure implementation with 53% auto-remediation coverage (16/30 functions, 236 tests)
- Complete EC2 auto-remediation testing (8/8 functions, 134 tests) with ASFF standardization patterns
- Complete RDS auto-remediation testing (7/7 controls, 88 tests) with comprehensive edge case coverage
- Established documentation-first testing methodology for efficient test development
- Added centralized test data management via fixtures/asff_data.py for consistent ASFF structures
- Implemented critical pytest module import isolation to prevent cross-contamination between test suites
- Enhanced testing documentation with LocalStack Docker integration and contributor guidelines
- Added bug handling protocol for test development to ensure proper production code review processes
- Comprehensive documentation added to core SOAR functions and auto-remediation components
- Testing strategy now serves as template for expanding coverage across all OpenSecOps repositories
Full Changelog: v2.1.2...v2.1.3
v2.1.0
- Added Lambda Layers for shared code (aws_utils and rds_remediation) to centralize cross-account functionality
- Reorganized all RDS autoremediations to leverage common code from the rds_remediation layer
- Fixed RDS.2 autoremediation to properly handle DB instances within Aurora clusters
- Fixed capitalization in RDS.9 and RDS.2 autoremediations to handle Security Hub ASFF format correctly
- Fixed access to parameter group fields using 'DbParameterGroups' instead of 'DBParameterGroups'
- Fixed access to cluster parameter group fields using 'DbClusterParameterGroups' instead of 'DBClusterParameterGroups'
- Enhanced RDS.9 autoremediation to handle DB cluster findings more reliably with API-based fallbacks
- Improved error handling in RDS.9 parameter group creation to prevent duplicate name conflicts
- Added unique suffix generation for RDS parameter group names to avoid name collisions
- Thoroughly tested RDS.9 autoremediation with PostgreSQL and Aurora PostgreSQL instances (standalone and in clusters)
Full Changelog: v2.0.1...v2.1.0