Skip to main content
View as MarkdownOpen in ChatGPTOpen in Claude

Identity Mesh — August 19, 2026

Release Date: August 19, 2026
Release Type: Mixed (New Features, Improvements, Bug Fixes, Security Fixes)

Overview​

This release introduces software inventory and compliance tracking for tenant-managed assets, adds self-service HR system onboarding, and resolves performance bottlenecks affecting large-scale software inventories.

  • SBOM/CBOM inventory — group assets by publisher or component, and trace a non-compliant component back through its dependency path.
  • HRIS onboarding — connect your HR system from the dashboard and complete the authorization without leaving Mesh.
  • Software inventory performance — pages that previously took over a minute to load now respond in under a second for tenants with tens of thousands of assets.

It also addresses entity-label validation logic and upgrades critical dependencies to resolve disclosed vulnerabilities.

New features​

SBOM and CBOM inventory with compliance tracking​

Administrators can now track software bills of materials and cryptographic bills of materials across managed endpoints and view compliance posture at a glance.

What you get:

  • Server-side grouping by publisher or component type, reducing client-side overhead for large inventories.
  • Publisher identity resolution for asset attribution and supply-chain visibility.
  • A dependency graph pruned to the paths that lead to a non-compliant component, so you see why something is non-compliant rather than the whole tree.
  • A per-organization switch for the dependency graph under Settings → General.

Why it matters: Organizations subject to software supply-chain mandates can see what they run and trace a non-compliant component back through its dependency path without exporting raw data to third-party tools.

HRIS onboarding​

You can now connect your HR system to Mesh directly from the dashboard.

What you get:

  • The connection flow runs inline in the dashboard, in Mesh branding — you authorize your HR system without leaving Mesh.
  • Tenant administrators can complete the connection themselves, without escalating to Axiad.
  • Once connected, HR records flow in alongside your other identity sources and appear in the connector list like any other connection.

Why it matters: your HR system is the authoritative record of who works for you and when they join or leave. Connecting it directly means joiner and leaver activity reaches Mesh without a manual export or a custom integration.

See Connect your HR system to set one up.

Improvements​

Timezone preference moved to Settings → General​

The timestamp display preference now lives with the organization it applies to, under Settings → General → Timezone. The separate Timestamp Display tab has been removed.

Why it matters: the setting is now scoped to the organization you are signed in to, so there is one place to set it and no chance of changing it for the wrong organization.

Bug fixes​

Correlation page went blank on an unrecognized status​

Opening the correlation page could render a blank screen when a record carried a status the page did not recognize. The page now handles unknown statuses instead of failing to render.

Software inventory timed out for large tenants​

Tenants managing tens of thousands of software assets experienced page-load failures exceeding 90 seconds when viewing the grouped asset list.

What changed:

  • Introduced a read model that pre-aggregates grouped asset data at the group grain, reducing query complexity from per-asset to per-group.
  • Load times dropped from 94.5 seconds to 31.4 seconds for typical large inventories, and from 85.9 seconds to 0.01–0.07 seconds when group counts are cached.

Why it matters: Administrators can now navigate software inventory pages reliably, even in environments with extensive endpoint fleets.

Entity-label validation rejected valid configurations​

The correlation engine incorrectly failed requests when entity labels were present but unmapped to a known schema, counting users instead of validating the label structure itself.

What changed:

  • Label validation now rejects only malformed labels; unmapped labels no longer trigger false negatives.

Why it matters: Custom labeling schemes no longer require pre-registration to avoid spurious validation errors during entity creation or update.

Security fixes​

  • GHSA-r277-6w6q-xmqw — updated the kin-openapi library used by the administrative API service.
  • CVE-2026-67213 — pinned the nanoid library to version 3.3.18 in the dashboard build.

Operator actions​

If this release modified a documented public API endpoint, re-validate any integrations that consume those endpoints against the current API reference. No action is required if you rely solely on the dashboard UI.