Deliver with Evidence: Safer Orchestration, Smarter Rollouts, and Scalable Processes (Release 25.3)

According to recent surveys, 31% of DevOps leaders said a lack of skilled resources is their biggest challenge while Legacy systems and infrastructure are a problem for 29% of DevOps leaders.

This is because organizations are required to manage excessive amounts of tools and fragmented processes, making performance and governance standards hard to measure and enforce. Digital.ai Release 25.3 introduces capabilities that consolidate toolchains, improve release decisions, and simplify operations. Together, they reduce sprawl, increase visibility, and align teams around outcomes:

• MCP Server (Beta) – An AI-based control plane to design, operate, and analyze releases with LLM prompts.
• Live Deployments – Improved RBAC controls and Git support for better visibility and security.
• Plugin & Integration Pack – AWS Secrets Manager, Dynatrace synthetic monitoring, SonarQube, GitLab, Octopus, ServiceNow, Confluence, and Tekton.

MCP Server (Beta)

Engineering teams need a single place to operate releases and diagnose failures. Release’s new “MCP Server” provides an AI-based engine to create, search, and manage releases; design templates; tune gates; and analyze failures—while executing existing Release features.

• Reduce drift and duplication with reusable, centrally managed policies for promotions, gates, and evidence, so teams inherit the same rules across folders and templates.
• Speed triage by asking natural-language questions (e.g., which promotions failed, which gate and why); MCP returns runs, artifacts, and logs and proposes next steps.
• Standardize configuration for tools and variables by scope (folder, template, release) so new projects adopt best practices automatically.
• Improve incident response and audits with a single surface to inspect failures, adjust gates, and record changes—turning scattered pipeline logic into consistent, governed operations.

Live Deployments

DevOps teams need a single place to see Kubernetes deployments and act on problems quickly. Digital.ai’s Live Deployments provides real-time visibility and control by aligning GitOps controllers with release governance.

• Ensure reliable, auditable Argo CD connectivity with explicit RBAC setup (service accounts, API tokens, webhook permissions) documented step by step.
• Manage Flux CD via container workflows that register Git sources, generate manifests/Helm, and expose sync/health as first-class signals.
• Correlate controller state to the governing release so you can filter by application/environment, compare success and frequency across stages, and take action without leaving Release.
• Reduce drift and speed incident response times with a single timeline of deployments, syncs, health changes, approvals, and tasks—turning Git and cluster state into policy inputs for hold/advance/rollback.

A look at the dashboard used to assess live deployments across multiple projects.

Plugin & Integration Pack

Evidence is commonly scattered across tools; secrets management and quality checks are manual; and CI/CD engines report status in different shapes, forcing teams to write custom code. Release 25.3 expands what you can automate inside the pipeline, bringing evidence and actions into one place, with the following integration updates:

• Tekton (Container Plugin): trigger and track pipelines in Kubernetes and drive Release flow based on run outcomes.
• AWS Secrets Manager (AWS Container Plugin): create, fetch, update, and delete secrets.
• Dynatrace: create synthetic monitors, run them on demand, and gate promotions on results.
• Confluence: add watchers, labels, and restrictions to keep stakeholders informed.
• Octopus Deploy: create releases against a pinned Git ref for config-as-code projects.
• ServiceNow: create/update CMDB CIs and records and enrich change requests with affected CIs.

User Interface Highlights

Improved user experiences allow for greater efficiency and simplicity. Navigation is faster, investigations are simpler, and guards like timeouts and risk signals are easy to review.

• Release list & calendar: more sorting options, lists of applied filters (title, owner, tags ALL/ANY, status, risk), and folder-level calendar views.
• Task Drawer history: filter by user/type, sort by time, investigate without leaving the task.
• Phases: color picker (HEX/RGB) and adaptive widths for deep nesting; clearer readability for complex plans.
• Risk & timeouts: enable/disable risk calculation in the UI; set timeouts for scripts, preconditions, and failure handlers.
• Runner Registry base URL: robust sub-path support and URL validation for private registries.

The Task Drawer now includes a “History” tab, providing a comprehensive view of the activity log for that task.

Releases can now be filtered by title, and the remaining filters have been reorganized for easier use.

Conclusion

A shared AI control plane (MCP), updated workflows, and expanded integrations reduce tool sprawl, improve visibility, and help teams meet business goals with confidence. By consolidating decision logic and automating the proof behind every change, you ship faster with fewer incidents, lower operational cost, and audit-ready records.

Review our release notes to discover how Digital.ai Release 25.3 can transform your software delivery process by providing AI-enabled orchestration, improved security and versioning, and improved integrations.