Harness Release Notes Summaries
Explore release notes from the last 30 days across the Harness Platform and modules.
INFO
Please review the full module release notes by selecting a module in the sidebar, or using the View full release notes links beside each module summary.
Looking for available features rather than recent changes? You can explore Beta and Limited GA features across Harness modules on the Feature Availability page.
Platform Release Notes
Platform
View full release notes →User Management & Security
- Ensured that all associated API Keys and Personal Access Tokens are automatically deleted when a user is removed from the system.
- Resolved an issue where User Group listings failed when reusable roles with identical identifiers but different scopes were assigned.
- Fixed an issue where users with partial view permissions received incorrect responses from the delete API.
APIs & Documentation
- Added request payload validation to the Role Assignment filter API endpoints to provide clearer error responses for invalid payloads.
- Improved the reliability and logging of the Purchase Credits API.
- Updated the API documentation for ending user impersonation sessions to accurately reflect its actual response behavior.
System Reliability
- Resolved a system race condition to improve overall platform stability.
- Restored a missing system configuration required for proper platform functionality.
Delegate
View full release notes →Security & Dependencies
- Addressed security vulnerabilities in the Netty library.
- Addressed security vulnerabilities in the GlassFish library.
Connectors & Integrations
- Improved Terraform Provider connector validation by using the connector's delegate selector instead of the executing delegate's selector, preventing conflicts when different delegates are used.
- Added support for Helm repositories using Open Container Initiative (OCI) registries from Google Artifact Registry.
- Resolved an issue where AWS connectors failed with an ambiguous 'Unknown error occurred' message.
Pipeline & Step Execution
- Updated pull request triggers to fail early with a clear error message if the associated Git connector lacks API access, preventing silent failures during commit data retrieval.
- Added runtime input support for destination and header routing configurations in pipeline steps.
Deployments & Scaling
- Fixed an issue where Amazon ECS scheduled scaling failed when start or end times included timezone offsets.
- Resolved a delegate selector collision that caused Helm Canary deployments to fail when using Git connectors pinned to different delegates.
Error Handling & Reliability
- Improved validation and error messaging to clearly indicate when an Amazon EKS cluster name is empty or invalid.
- Fixed an issue where the Terraform Cloud Apply step intermittently retried despite succeeding, caused by temporary log streaming disconnections.
- Fixed a potential system error during deployment error reporting when password resolution returns no value.
Self-Managed Enterprise Edition
View full release notes →Security & Access Management
- Added support for passing delegate details as session tags in OIDC tokens to enforce identity policies based on which delegate runs a task.
- Added OIDC authentication support for AWS GovCloud regions.
- Resolved an issue where token rotation failed for users with specific organization or project permissions.
- Prevented secret values from being logged in plaintext in delegate logs during connection tests.
- Ensured policy warnings are correctly displayed in the user interface.
Build & Test Optimization
- Enabled versioned Harness CLI binaries in build runners to improve stability and rollback capabilities.
- Improved test splitting accuracy using historical average durations for more balanced parallelism.
- Added support for caching NuGet dependencies based on environment variables.
- Added the ability to override Azure-specific storage fields in cache settings.
Pipelines & Git Integration
- Added a validation API endpoint to check pipeline configuration changes in Git before committing.
- Added support for referencing pipeline configurations and input sets using Git tags during trigger-based executions.
- Introduced a central page to view, filter, and manage queued pipeline executions.
Deployment & Cloud Integrations
- Added direct deployment support to Azure Container Apps with progressive traffic shifting and automated rollbacks.
- Resolved an intermittent delay when downloading configuration files from Amazon S3.
- Fixed validation issues when pulling Helm charts from Amazon ECR.
Platform Performance & Administration
- Optimized permission checks in the environments list API to prevent timeouts and improve performance.
- Made module name matching case-insensitive for easier configuration.
- Fixed an incorrect warning message when deleting organization- or project-level users.
- Corrected the sorting of user groups in the interface.
- Prevented creation of entities using reserved system identifiers to avoid update issues.
AI for DevOps & Automation
Continuous Delivery & GitOps
View full release notes →Security & Access Control
- Added OIDC authentication support to the Artifactory connector for credential-free federated access.
- Added environment-specific session tags to OIDC tokens for AWS connectors, enabling targeted IAM policy enforcement.
- Introduced OIDC Delegate Selectors for AWS Connectors to pass delegate details as session tags, improving secret isolation.
- Fixed an issue where the pipeline ID was missing from the OIDC sub claim for containerized step groups.
- Fixed a permission loss issue that caused Google Cloud Functions Gen2 rollbacks to fail.
Deployment & Pipeline Execution
- Added a 'Preserve Directory Structure' option to the Command step to prevent file overwrites during configuration copies.
- Added support for additional command flags, such as server-side apply, in the Kubernetes Dry Run step.
- Fixed an issue where concurrent custom artifact polling failed due to shared script file paths.
- Fixed an issue where a rollback failure prevented subsequent steps configured with 'Always' conditions from running.
- Fixed an issue where the Terraform Cloud Apply step was incorrectly retried after a successful execution due to log streaming loss.
Git Integration & Templates
- PR triggers now require API access to be enabled on their linked Git connector to execute pipelines.
- Improved performance of the GitOps UpdateReleaseRepo step under high concurrency by addressing GitHub token lock issues.
- Fixed a manifest compilation failure during dry runs when applying large Kubernetes Custom Resource Definitions from GitHub.
- Fixed an issue that prevented users from importing templates from Bitbucket.
- Fixed an issue that prevented users from creating templates directly from a pipeline.
Governance & Traffic Management
- Enhanced Harness AI capabilities to generate and describe Open Policy Agent (OPA) policies in plain language.
- Added configurable matching logic (AND/OR) for Istio traffic routing step rules.
- Added a setting to allow users without approval permissions to view approval step details during pipeline execution.
User Interface & Developer Experience
- Added a new dry-run API endpoint to validate pipeline YAML, templates, and policies before committing changes to Git.
- Fixed an issue where viewing parallel environments in the visual editor incorrectly prompted users with unsaved changes.
Continuous Integration
View full release notes →Cloud & Network Connectivity
- Added the ability to configure network egress restrictions for cloud build virtual machines directly from the user interface.
- Added support for using custom container images from private registries in cloud builds.
- Improved private network connectivity reliability by automatically configuring selectors for private connections.
- Ensured connector validation tasks successfully route through cloud infrastructure for private network setups.
Build & Caching Performance
- Added Azure Blob Storage as a supported storage backend for Docker Layer Caching and Build Cache operations.
- Optimized build initialization on Kubernetes infrastructure by reducing payload size, resulting in faster and more reliable build startup.
- Resolved cache key collisions that occurred when running parallel build and push steps across different platforms.
- Fixed an issue that prevented custom access control configurations from applying to cached objects.
- Fixed a conflict where exporting AWS credentials as output variables caused Cache Intelligence steps to fail.
Security & Stability
- Resolved security vulnerabilities by upgrading the command-line interface and patching the cache image.
- Fixed an issue where log sanitization patterns were not consistently applied to container steps running on Kubernetes.
- Improved log streaming resilience by gracefully handling redirect and error responses from the log service.
- Upgraded command-line binaries to versioned instances to improve management and rollback capabilities.
Pipeline Execution & Management
- Added a test management dashboard that displays the health and history of all tests, including flaky and quarantined tests.
- Resolved a permission-denied error encountered during Git operations in Self-Managed Platform environments.
- Fixed a storage driver conflict that caused build and push steps to fail with file system errors.
- Fixed an issue where referencing unresolved environment variables added empty values to pipeline execution tags.
- Fixed an issue where using image digest references instead of tags triggered false warnings about outdated image versions.
Artifact Registry
View full release notes →Security & Policy Control
- Block artifacts that exceed configured open-source risk thresholds, factoring in maintenance status, vulnerabilities, and community health.
- Block artifact versions flagged as malicious in threat intelligence databases.
Lifecycle & Cleanup
- Configure cleanup rules to soft-delete artifact versions based on age, version count, or name patterns, while keeping them recoverable.
- Set retention rules to protect specific artifacts from deletion attempts, overriding cleanup rules and manual deletion.
- Preview which artifact versions a rule would affect using dry-run mode before enabling it.
Governance & Monitoring
- Create and apply rules at the account, organization, or project level across multiple registries.
- Track rule executions with detailed logs showing which artifact versions were deleted or retained.
- Receive centralized notifications when cleanup rules or dry-run executions complete.
AI for Testing & Resilience
Feature Management & Experimentation
View full release notes →Pipeline Integrations
- Added support for integrating Feature Management and Experimentation directly into Harness Pipelines to orchestrate and automate your feature releases.
API & Developer Tools
- Introduced a new Admin API endpoint that allows developers to perform full updates on feature flag definitions within an environment.
Feature Flags
View full release notes →Documentation & Reference
- Detailed updates and release notes for this period are available on the Feature Flags documentation page.
Chaos Engineering
View full release notes →Security & Compliance
- Resolved high-severity security vulnerabilities in multiple chaos components, including the source probe, log watcher, and DDCR faults.
- Added audit trail support for the image registry, including visual side-by-side YAML differences.
- Enabled the propagation of infrastructure security contexts directly to experiment and helper pods.
Experiment & Template Management
- Added support for Linux and Windows Chaos Experiment Templates as part of the pipeline Chaos Step template.
- Introduced conditional execution for faults, probes, and actions, giving you control over when steps run based on custom conditions.
- Fixed an issue where optional parameters were treated as mandatory when set as runtime inputs in an experiment template.
Infrastructure & Execution
- Added remote Kubernetes execution support for Linux and Windows experiments, enabling you to run probes, actions, and faults against a customizable default Kubernetes infrastructure.
- Improved handling for helper pods when they enter a Pending or Failed status.
- Fixed an issue where the worker count was not correctly passed when initiating a Kubernetes load test from the list page.
- Updated internal chaos execution components to improve overall stability.
Probes & Monitoring
- Optimized handling for probe timeouts, durations, and retries.
- Fixed an issue where probes could run longer than their configured timeout duration.
- Removed legacy Prometheus, Datadog, and Dynatrace probe types from Kubernetes to streamline probe management.
User Experience & Insights
- Added a new risk insights landing page and dedicated navigation routes.
- Added the ability to copy output variables directly from the experiment timeline view node.
- Fixed an issue where load test dashboard graphs showed visual gaps due to dropped metric snapshots during transient gateway failures.
AI Test Automation
View full release notes →Playwright Automation Support
- Introduced the Playwright execution engine for running automated tests.
- Added support for drag-and-drop interactions in Playwright tests.
- Added clipboard support for copy-and-paste operations within Playwright sessions.
Test Authoring & Management
- Enabled generating and downloading test cases written in natural language.
- Prevented the creation of duplicate test suites to keep test repositories clean.
- Fixed an issue where creating tasks from test steps did not correctly enable nested tasks.
- Configured general tasks to execute exclusively as manual actions.
User Interface & Session Management
- Improved the handling and visibility of session expiry warnings and timeouts.
- Blocked manual click commands during automated test execution to prevent process interruption.
- Optimized code formatting and presentation for session components.
Last updated Jun 24, 2026
On-Call & Incident Routing
- Selectively import on-call configurations, including users, teams, schedules, and escalation policies, from PagerDuty, OpsGenie, and xMatters with a preview step and background sync tracking.
- Target specific rotations within a schedule when configuring escalation policies, enabling more precise incident routing.
Runbooks & Automation
- Duplicate existing runbooks to quickly create variations with all configurations and metadata preserved.
- Use Common Expression Language (CEL) to write advanced dynamic conditions across alerts, runbook triggers, and webhooks with built-in validation.
Collaboration & Integrations
- Mirror incident timelines bidirectionally with Google Chat and automate channel management using native runbook actions.
- View incident summaries in Slack via a dedicated modal dialog to reduce channel noise.
AI for Security & Compliance
Security Testing Orchestration
View full release notes →Jira Integration Updates
- Enhanced Jira integration to include vulnerability summaries and detailed occurrence information directly inside Jira tickets.
- Added support for creating Jira tickets for security issues found in non-baseline targets, such as feature branches and pull requests.
Policy & Exemption Management
- Introduced customizable exemption durations, giving requesters and approvers more flexibility and control when managing exemption requests.
User Interface & Filtering
- Fixed an issue where searching by Issue ID did not return results for remediated or exempted issues on the Issues page.
- Resolved an issue where stage names were not displayed alongside step names in the Vulnerabilities tab filter.
Supply Chain Security
View full release notes →Access Control & Permissions
- Introduced granular Role-Based Access Control (RBAC) across Account, Organization, and Project scopes, allowing precise control over Supply Chain Security workflows.
- Enforced permission limits for view-only roles, preventing unauthorized write operations on Supply Chain Security resources.
Security & Risk Management
- Added detection for malicious packages and typosquatting risks in open-source dependencies, enabling risk filtering and pipeline policy enforcement.
- Introduced automated License Family Classification in Software Bill of Materials (SBOM) to simplify license risk assessments and support custom mapping and policy-based enforcement.
Integrations & Workflows
- Added support for discovering, onboarding, and managing Bitbucket repositories directly within Supply Chain Security.
- Resolved an issue where Software Bill of Materials (SBOM) orchestration steps failed during upload due to misconfigured custom settings.
AI for Cost & Optimization
Cloud Cost Management
View full release notes →Analytics & Cost Insights
- Refined preference handling in the Advanced Filters drawer to disable filtered-out providers in Cost Explorer.
- Updated the Overview anomalies section to accurately reflect your configured cost types instead of defaulting to standard cost types.
- Enhanced anomaly calculations to aggregate and display both resource-level and cost category-level active anomalies on the Overview and Perspective screens.
- Fixed an issue where net-amortized cost savings for node pool recommendations displayed inflated values, ensuring they align with monthly cost metrics.
Connectors & Settings
- Added a guided 3-step setup wizard with real-time connectivity testing to securely connect OpenAI and Anthropic accounts.
- Introduced a Cost Settings tab within Asset Governance, unifying alert, preference, and provider-specific cost configurations into a single settings drawer.
- Resolved an issue where the Anomalies Metrics API ignored deleted or invalid perspective filters, ensuring correct empty results are returned.
AI DLC Insights
View full release notes →Documentation Updates
- Refer to the dedicated AI DLC Insights release notes page for the latest updates and documentation.