Skip to Content
Getting StartedWhat's New

What’s New — Recent Enhancements

This page summarizes recent product enhancements so you can get the most out of the latest AtlasAI release. Use it to understand new behavior, improved workflows, and where the product is headed.

UI & Experience

Traces: Waterfall and flame graph

  • What: On Traces, open any trace to switch between Waterfall (timeline) and Flame graph (span depth vs duration). The flame view uses the same span data as the waterfall and supports clicking a span to focus details.
  • Why: Faster visual scanning of hotspots and nested work without losing the detailed waterfall when you need it.
  • When to use: Any distributed trace with stored spans in the tenant database. See Traces.

Command palette

  • What: A quick-action palette (keyboard shortcut or menu) lets you jump to any major area — Incidents, Runbooks, Dashboards, Data Sources, Settings — and run common actions without using the sidebar.
  • Why: Faster navigation and fewer clicks, especially for power users and on-call workflows.
  • When to use: Anytime you know where you want to go; use the shortcut to open the palette and type the destination or action. See Using the interface.

Responsive sidebar and layout

  • What: The main navigation sidebar adapts to screen size: full sidebar on desktop, collapsible or hamburger menu on smaller screens so you can focus on content.
  • Why: Better use of space on laptops and tablets; consistent experience across devices.
  • When: Use the collapse/expand control when you need more horizontal space for dashboards or incident details.

Forms and validation

  • What: Forms across the product (incident creation, runbook edit, policy creation, user invite) use consistent validation and clear error messages. Invalid fields are highlighted and explained.
  • Why: Fewer submission errors and faster correction when something is wrong.
  • When: Whenever you create or edit entities; fix any highlighted errors before submitting.

Charts and visualizations

  • What: Time-series charts, topology graphs, and data tables include improved labeling, legends, and keyboard/screen-reader support so everyone can read and navigate them.
  • Why: Accessibility and clarity for all users; better decisions from clearer data.
  • When: Use charts in Dashboards and incident evidence; use topology in RCA and impact analysis.

Error handling

  • What: If a section of the app fails to load, you see a clear error message and recovery options (e.g. retry or go back) instead of a blank screen. Errors are contained so the rest of the page remains usable.
  • Why: You always know what went wrong and what you can do next.
  • When: If you see an error boundary message, use Retry or navigate away and try again; report persistent errors to your admin.

Lists and pagination

Alerts and content packs

  • What: Large lists (e.g. alerts, content packs) are paginated. You see a limited number of items per page with Previous / Next (and optionally page size) so the UI stays responsive.
  • Why: Reliable performance with large datasets; no long freezes or timeouts.
  • When: Use pagination controls when browsing alerts or content packs; use filters and search to narrow results when possible.

Incidents and other lists

  • What: Incident lists and other list views support pagination and consistent limits (default and maximum page size) so the API and UI behave predictably under load.
  • Why: Same as above — stability and predictable behavior at scale.
  • When: When you have many incidents or items, use the list controls to move through pages and adjust page size if the product allows it.

Security and reliability

Rate limiting (login and provisioning)

  • What: Sign-in and sensitive flows (e.g. password reset, tenant provisioning) are rate-limited to prevent abuse (e.g. brute-force or enumeration). After too many attempts in a short window, further attempts are temporarily blocked with a clear message.
  • Why: Protects your account and the platform from automated attacks.
  • When: If you see a rate-limit message, wait a few minutes and try again; use Forgot password if you’ve forgotten your password instead of repeated guesses.

Password reset

  • What: Forgot password and reset password flows are available on the login page. Reset links are time-limited and single-use; the system does not reveal whether an email exists to avoid enumeration.
  • Why: Secure self-service recovery without involving support for every forgotten password.
  • When: Use Forgot password on the login page when you don’t remember your password; complete the reset in the same browser session when possible. See Authentication & user management.

API consistency and errors

  • What: APIs return consistent JSON shapes for success and errors. Error responses include a message and often a trace ID for support; they do not expose internal stack traces in production.
  • Why: Easier integration, debugging, and support; no leakage of sensitive implementation details.
  • When: When integrating with AtlasAI (webhooks, API clients), rely on the documented response format and use the trace ID when reporting issues.

Automation and reporting (Tenant Plane)

Workflows — definitions, canvas, and instances

  • What: Workflows (/workflows) now centers on Definitions and Instances. You build steps in Steps (runbook picker and drag-to-reorder), Canvas (visual graph), or JSON (full step types: runbook, approval, condition, notification, delay, generate_report). Run starts an execution job; Last run result shows output. Instances lists runs with Approve / Reject for pending approval steps.
  • Why: One place to design operational automation chains and human gates without confusing them with ITSM catalog approval definitions.
  • When: After runbooks exist; use Workflow builder (step-by-step) for a checklist. Singular /workflow redirects to /workflows.

Report Builder — definitions, runs, schedules

  • What: Reports (/reports) is the Report Builder: Definitions and Runs tabs, + New Report with type, format (HTML/CSV/JSON), data sources, and optional Daily / Weekly / Monthly schedule (UTC). Run opens HTML in a tab or downloads CSV/JSON. Header link Dashboards goes to saved report dashboards.
  • Why: Self-service operational and executive reports from tenant data with optional scheduling.
  • When: Use Reports for the full step-by-step; use Report dashboards for named layout records under /reports/dashboards.

Data and operations

Retention and cleanup

  • What: Retention policies control how long different types of data (e.g. logs, metrics, incidents) are kept. A retention worker runs on a schedule to apply these policies and free storage.
  • Why: Compliance, cost control, and predictable storage growth.
  • When: Configure retention in Settings or via your deployment; see your admin or On-prem installation for where retention is configured.

Type safety and reliability

  • What: The platform codebase has been hardened with stricter typing and validation so that APIs and the UI behave consistently and fail predictably (e.g. validated IDs, pagination limits).
  • Why: Fewer runtime surprises and better long-term reliability.
  • When: As a user you benefit automatically; when integrating, use validated parameters and respect documented limits (e.g. max page size).

Where to go next

Architecture OverviewDeployment models (TP)