Release notes — July 6, 2026

Summary

Build 20260706.1. A major release focused on data-driven testing, enterprise collaboration and access control, and a smarter AI foundation. You can now run any test or workflow across entire datasets, manage teams with workspaces, SSO, and fine-grained roles, and benefit from a test-generation engine that learns from your API and produces more coherent, business-aware tests.

Improvements

• Data-driven testing, built in: run any test or workflow across an entire dataset — one test now executes across N rows automatically. Manage reusable data sources with the new Dataset Manager and inline data grid.
• Datasets in Project Settings: create, edit, version, and import/export datasets directly from Project Settings → Data, with support for CSV, Excel, JSON, XML, YAML, databases, APIs, and scripts.
• Smarter data-aware generation: test generation now uses all the rows you upload instead of just the first, and automatically produces data-driven iterations — including negative variants from INVALID_* columns.
• Dynamic value tokens: a greatly expanded library of dynamic tokens (dates, UUIDs, random values, and more) that resolve inside data-table cells and anywhere values are used.
• Workspaces and per-project roles: organize projects under workspaces and assign members granular, per-project roles for cleaner team access.
• Enterprise single sign-on: SSO via OIDC, SAML, and Microsoft Entra, with just-in-time user provisioning and group-to-role mapping.
• Group-based access, API keys, and webhooks: manage access by group, issue scoped platform API keys, and deliver outbound webhooks with retry and redelivery.
• External secret managers: pull credentials from external secret-manager integrations instead of storing them inline.
• Comments and collaboration: leave comments directly on tests and endpoints to coordinate with your team.
• Fine-grained permission enforcement: custom roles are now consistently enforced across the platform.
• AI Learning Intelligence: the platform now learns from your API's real behavior to generate more accurate expected results, with continuous curation and quality evaluation behind the scenes.
• Fix-with-AI and self-healing: get AI-suggested fixes for failing tests, with optional self-healing — proposals only, so product bugs are never silently "fixed" away.
• More coherent test generation: an overhauled coverage engine produces behavioral, business-logic-aware tests instead of shallow, repetitive ones — with broader protocol coverage for GraphQL and SOAP.
• Test Workbench: a new split-view surface to edit, run, and see results side by side — and pop any test out into its own window.
• Request assertions: assert on the outgoing request itself, guarded behind an opt-in toggle.
• Faster reports: removed redundant per-test data fetching for noticeably quicker high-level and detailed reports.

Fixes

• Postman import: multi-test Postman collections now import correctly.
• Desktop app stability on Windows: resolved a startup crash in the compiled local runner related to secure credential storage.
• Licensing: fixed a false license-revocation edge case and hardened license handling.
• Environments: a newly set default environment now applies to runs immediately.
• Test execution: corrected path-parameter substitution and made status assertions more flexible for HTTP-semantics tests.
• Reporting: per-test data-driven reports are now correctly scoped to that test's own iterations.
• Broad stability pass: resolved 30+ additional confirmed issues across mock caching, auth profiles, the activity feed, modals, RBAC gating, AI settings, and more.

Known issues

• None reported at this time.

Related links

• Browse all release notes
• Help Center
• Product documentation