# February 2026
{% updates format="full" %}
{% update date="2026-02-25" %}
## Improved reliability, clarity, and authentication flow
#### ⚡ Improvements
**Cloud import agent has a new look**
The agent's output panel in codification workflows now groups messages by step and state (working/success/failed). We also added collapsible code panels and clearer progress indicators to help review long-running jobs.
**What does it mean for you?**
You get better visual organization and faster insight into job progress or errors during agent execution.
***
**Uninstall worker supports multiple deploy types**
Uninstall jobs now pass all deploy types correctly to the backend API, ensuring consistent behavior between environment TTL expirations and manual uninstall flows.
**What does it mean for you?**
Automated cleanup and lifecycle management for multi-deploy blueprints are now more predictable and reliable across environments.
***
**Blueprint publish validation for child outputs**
When publishing a parent blueprint, the system now validates that all referenced child outputs exist. Missing outputs trigger early, clear validation errors.
**What does it mean for you?**
You’ll catch missing dependencies during publishing instead of facing runtime failures during deployment.
***
**Clearer feedback when uninstall Is blocked**
When you try to uninstall while a run is still active, the modal now shows a clear inline banner instead of a generic toast message.
**What does it mean for you?**
You’ll immediately understand why an uninstall can’t proceed, reducing confusion during active environment changes.
***
**Cleaner runner logs and smarter retry logic**
Initialization now fails early for unsupported OpenTofu versions and stops retrying non-recoverable errors. We also removed unnecessary “No active git configuration” messages during runner startup.
**What does it mean for you?**
Runner logs are quieter and more actionable, helping you diagnose setup issues faster and avoid blind retries.
***
**Terminology and microcopy consistency updates**
Updated various interface labels and copy for consistency (e.g., “Target collection” → “Collection”, “Latest runs” → “Run history”). Improved capitalization and creation guidance for collections and account settings.
**What does it mean for you?**
A more consistent and polished experience across the interface, reducing naming confusion between flows.
***
**Blueprint creation terminology: “Artifact”**
We’ve standardized on the term **artifact** throughout the blueprint creation flow, replacing older “package” references.
**What does it mean for you?**
Terminology now aligns with the platform’s current object model, clarifying the relationship between blueprints and published artifacts.
***
**Azure tooltip link update**
The Azure connection tooltip now points to the correct (hidden) documentation page for guided setup.
**What does it mean for you?**
You’ll reach the right instructions faster when setting up or debugging Azure integrations.
***
#### 🐛 Bug fixes
**Fixed incorrect error during CLI uninstall**
Resolved a misleading error message shown when uninstalling environments via CLI. The uninstall now starts correctly or provides a relevant error prompt.
***
**Resolved allowed values type conversion issue**
Fixed a mismatch between default and allowed values (string vs. number) in the blueprint editor that caused publish failures during manual hydration.
***
**Fixed task controller race condition**
Addressed an orchestration race condition that triggered intermittent “object has been modified” errors and redundant alerting.
{% endupdate %}
{% update date="2026-02-18" %}
## More automated environments, fewer manual steps
#### ⚡ Improvements
**Improved run visibility across the plan page and environment API**
The plan page now displays run type indicators (incremental, auto-approve, auto-apply) and supports enhanced stage filtering. Environment detail responses also include execution flags for incremental and force installs.
**What this means for you:**\
You get clearer insight into how runs are being executed, both in the UI and through the API.
***
**More reliable retries for package downloads**
Package downloads now retry correctly on transient network failures, such as connection resets.
**What this means for you:**\
You’re less likely to experience failed runs due to intermittent network issues, especially in CI/CD environments.
***
**CLI terminology updated from deployment to environment**
The CLI now consistently uses “Environment” instead of “Deployment” across commands, models, and documentation.
**What this means for you:**\
Terminology is clearer and better aligned with the Bluebricks object model, reducing confusion across tools and APIs.
***
**Improved log readability in the UI**
We’ve upgraded the logs drawer and added support for UNIX color codes in the agent sidebar.
**What this means for you:**\
Logs are easier to scan and interpret, especially during complex runs with colored output.
***
**Cleaner Secrets tab experience when encryption keys are missing**
Opening the Secrets tab no longer shows an immediate error toast when no public key is configured. Guidance remains available via tooltip.
**What this means for you:**\
A less noisy UI while still clearly indicating what’s required for encrypted secrets.
***
**Blueprint creation CTA updated for clarity**
The primary action in the blueprint creation flow now reads “Create and publish.”
**What this means for you:**\
It’s clearer that creating a blueprint will immediately publish it as part of the workflow.
***
**Backend permissions aligned with Environment model**
Backend routes now rely on environment permissions instead of legacy deployment permissions.
**What this means for you:**\
Access control is more consistent with the current data model, helping reduce ambiguity as the platform evolves.
***
#### 🐛 Bug fixes
**Fixed CLI publish hang in non-interactive CI/CD workflows**
Resolved an issue where `blueprint publish --non-interactive` could hang indefinitely if the version already existed.
**What this means for you:**\
Your CI/CD pipelines no longer get blocked by stalled publish steps.
***
**Runner no longer loops endlessly on unauthorized Kubernetes access**
The Helm executor now detects Kubernetes authorization failures and exits appropriately instead of retrying indefinitely.
**What this means for you:**\
Kubernetes permission issues surface immediately instead of causing stuck jobs.
***
**Deployment creation no longer blocked by unused collection property validation**
Fixed an issue where deployments could fail due to validation of collection properties that were not actually referenced.
**What this means for you:**\
You won’t encounter incorrect missing-output errors when creating deployments.
***
**Auto-approve deployments now start with the correct run type**
Selecting auto-approve no longer triggers auto-apply behavior.
**What this means for you:**\
Deployments run with the intended approval mode, ensuring predictable execution.
***
**Azure Discovery details endpoint is more reliable for expandable resources**
Improved Azure resource detail lookups to handle inconsistent table mappings and ID casing.
**What this means for you:**\
You’ll see fewer errors when exploring expandable Azure resources in Discovery.
***
**Fixed occasional logout redirect to org selection screen**
Resolved an issue where users could be redirected to the org selection screen unexpectedly.
**What this means for you:**\
Session behavior is more stable and predictable.
***
**Blueprint page header spacing is now consistent**
Corrected a padding inconsistency in blueprint header blocks.
**What this means for you:**\
The blueprint page layout is cleaner and more visually consistent.
{% endupdate %}
{% update date="2026-02-11" %}
## Deployment & Discovery Enhancements
**🚀 New features**
**Azure discovery & Terraform integration**
We now support Azure Discovery end to end. You can connect Azure cloud accounts, see Azure resources directly in the Graph, and jump straight to Azure Terraform resources through direct hyperlinks.
**What does it mean for you?**\
You get a smoother multi-cloud experience, clearer visibility into your Azure infrastructure, and faster navigation between Bluebricks and your Terraform-managed resources.
***
**End-to-end auto-apply deployment mode**
We now support AutoApply deployments across the full stack. When policies allow, you can automatically apply changes, skipping the planning stages to substantially reduce runtime duration and enable faster, fire-and-forget workflows.
**What does it mean for you?**\
You spend less time waiting and dealing with manual approvals, making repeatable deployments significantly faster in trusted environments.
***
**Cron-based repeatable job scheduling**
We've added support for cron-based scheduled jobs, so you can automate recurring infrastructure and environment workflows.
**What does it mean for you?**\
You can easily schedule routine tasks like maintenance, drift detection, or timed environment actions without manual intervention.
***
**Environment scheduler APIs**
We introduced new APIs that allow environments to be scheduled, managed, and automated more effectively.
**What does it mean for you?**\
You get greater control and flexibility when integrating lifecycle automation into your workflows.
***
**⚡ Improvements**
**Simplified Graph canvas & resource filtering**
We've streamlined the Graph canvas with cleaner interactions, improved filtering, and a smoother collection creation flow.
**What does it mean for you?**\
You can explore and navigate complex infrastructure faster and with less friction.
***
**Improved Graph resource detail resolution**
We've improved how resource details are resolved in the Graph based on type and ARN values.
**What does it mean for you?**\
You see more accurate metadata and clearer context when inspecting resources.
***
**Deployment TTL visibility**
Deployments now clearly display their time-to-live (TTL), so you can see exactly when an environment is scheduled to expire.
**What does it mean for you?**\
You can better plan around temporary environments and avoid unexpected teardown.
***
**Improved logs for debugging and clarity**
We now preserve additional log fields and have cleaned up deployment output to reduce redundant messages and better reflect severity levels.
**What does it mean for you?**\
You get richer troubleshooting context with less noise.
***
**Stronger validation for configuration inputs**
We've tightened validation across configuration inputs. We now prevent invalid whitespace in subdirectory fields and reject invalid lowercase prefixes in output values.
**What does it mean for you?**\
You catch configuration mistakes earlier and avoid preventable deployment errors.
***
**Blueprint editing flexibility**
You can now edit more blueprint fields, with only package ID and technology remaining locked.
**What does it mean for you?**\
You have more flexibility when iterating on blueprint definitions without unnecessary restrictions.
***
**Environment flow UX polish**
We've refined the deployment creation flow so the "Environment name" field stays focused throughout the process.
**What does it mean for you?**\
You can move through environment setup more smoothly and with fewer interruptions.
***
**Improved Git credential handling for private repos**
We improved how we resolve and manage GitHub credentials for private repositories, including stronger org validation and credential handling.
**What does it mean for you?**\
You experience fewer authentication failures when working with private, source-based packages.
***
**Cloud Discovery reliability improvements**
We increased Cloud Discovery reliability with improved redundancy and availability protections.
**What does it mean for you?**\
You benefit from more stable discovery workflows with reduced downtime risk.
***
**Response structure alignment across agents**
We've standardized how agent responses are structured across workflows.
**What does it mean for you?**\
You get more predictable, readable outputs when working across different agents and automation flows.
***
**🐛 Bug fixes**
**Correct identity display in audit fields**
We fixed an issue where `created_by` showed a machine identity instead of the API key identity during deployment and environment creation.
**What does it mean for you?**\
You get clearer attribution and more accurate audit trails.
***
**Job scheduler stability fixes**
We resolved issues in the job scheduler APIs related to enabling and disabling schedulers.
**What does it mean for you?**\
Your scheduled automation runs more reliably with fewer interruptions.
***
**Execution and retry reliability**
We fixed multiple execution issues, including stale Terraform plan reuse after failed apply, log streaming timing mismatches, and uninstall handling for removed source-based artifacts.
**What does it mean for you?**\
Your deployments are more dependable, especially during retries and cleanup scenarios.
***
**Prevented recursive package directory creation**
We fixed an issue where invalid nested directory structures could be created during execution.
**What does it mean for you?**\
You get cleaner execution environments and fewer filesystem-related failures.
***
**Fixed Kubernetes unauthorized infinite loop in Helm deployments**
We resolved an issue where Helm deployments could enter an infinite loop after a Kubernetes authorization failure.
**What does it mean for you?**\
Failures surface faster and recovery is safer when cluster permissions are misconfigured.
***
**Graph drawer header visibility restored**
We fixed an issue where the Graph drawer header was not visible.
**What does it mean for you?**\
You get a consistent and fully usable Graph experience.
***
**"Deployments" are now called "environments"**
We fixed issues with semantic value references during deployment-to-environment renaming.
**What does it mean for you?**\
You avoid broken references and maintain stability when renaming infrastructure entities.
***
**Cleaner production logging output**
We removed unnecessary trace details from production CLI logs.
**What does it mean for you?**\
You see clearer, more actionable logs in production environments.
{% endupdate %}
{% update date="2026-02-04" %}
## Faster automation & graph usability improvements
**🚀 New features**
**Run AutoApply**
We introduced support for running deployments with `--auto-apply`, enabling a direct apply workflow. When appropriate, this skips the traditional plan and approval stages while keeping existing commands unchanged.
**What does it mean for you?**\
You can move from intent to execution much faster. For trusted workflows, you reduce runtime duration and eliminate unnecessary approval friction.
***
**⚡ Improvements**
**JSON references in inputs**
We now allow `$` references inside JSON-based inputs, so you can reuse outputs and resolved values directly within JSON payloads.
**What does it mean for you?**\
You gain more flexibility when composing inputs, with less duplication and cleaner configuration logic.
***
**Modernized job scheduling architecture**
We've improved job scheduling reliability and added support for cron-based repeatable jobs, including scheduled environment uninstalls.
**What does it mean for you?**\
You benefit from more reliable, observable, and scalable automation, especially in multi-instance environments.
***
**Increased resource limits for Generic Artifacts**
We increased resource limits for Generic Artifact executions to reduce failures during heavier workloads.
**What does it mean for you?**\
You experience fewer unexpected runtime failures during heavier deployments.
***
**More resilient uninstall flows**
We improved dependency handling to prevent uninstall deadlocks when some packages were not previously applied.
**What does it mean for you?**\
Uninstall operations are less likely to get stuck, even in partial or edge-case deployment states.
***
**Clearer Graph call to action**
We renamed the Resource Discovery CTA from "Create deployment" to "Import" to better reflect the codification flow.
**What does it mean for you?**\
The Graph workflow is clearer and more intuitive when importing discovered resources.
***
**Plan page simplification**
We removed the "View parent blueprints" option from package menus on the Plan page to streamline actions.
**What does it mean for you?**\
A cleaner Plan experience with fewer distracting or low-value actions.
***
**🐛 Bug fixes**
**Prevented webapp crash from null nodes**
We fixed a runtime crash that could occur when the UI received null node data.
**What does it mean for you?**\
You're less likely to encounter blank screens or unexpected errors.
***
**Improved resource selection behavior**
Clearing the search field in resource selection no longer resets your chosen resources.
**What does it mean for you?**\
You can refine your search without losing previous selections.
***
**Correct region validation in blueprint editor**
We fixed a validation issue where region inputs with allowed values were incorrectly marked as required in certain conflict scenarios.
**What does it mean for you?**\
Validation behaves correctly and avoids unnecessary friction in blueprint editing.
***
**Source package redeploy consistency**
We fixed an issue where updated props for source-based artifacts were not reflected properly after redeploy.
**What does it mean for you?**\
Schema updates, dependency ordering, and publishing behavior now remain consistent after redeployment.
***
**Deploy flow to existing deployment fixed**
We resolved an issue that prevented deploying a newly published blueprint version to an existing deployment without resetting the dropdown.
**What does it mean for you?**\
You can deploy new versions to existing environments more smoothly.
{% endupdate %}
{% endupdates %}
---
# Agent Instructions: Querying This Documentation
If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question.
Perform an HTTP GET request on the current page URL with the `ask` query parameter:
```
GET https://bluebricks.co/docs/changelog/february-2026.md?ask=
```
The question should be specific, self-contained, and written in natural language.
The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.
Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.