Testmo’s API has always been a reliable way to read your testing data — pull test cases, runs, results, milestones, and automation history into your own tools and dashboards. With this release, it becomes a two-way street. You can now create and manage milestones and record test results directly through the API, closing the loop between Testmo and the systems you build around it.

This is the second phase of our API expansion, and it means the workflows you used to run by hand — or work around with scripts — can now be driven programmatically, end to end.

Manage milestones through the API

Milestones are how you organise releases, sprints, and test cycles in Testmo. Now you can manage them without leaving your pipeline:

• Create a milestone with full metadata — name, milestone type, start date, due date, and description
• Update any of those details as a release evolves
• Delete a milestone cleanly when it’s no longer needed

Deletion is safe by design. When you remove a milestone, Testmo automatically unlinks its test runs, sessions, and automation runs, clears their associations, and reparents any sub-milestones — so nothing is orphaned and your history stays intact.

The obvious win here is release automation. When your CI/CD pipeline kicks off a release, it can now create the matching Testmo milestone automatically, wire your runs to it, and tidy up afterwards — no manual setup, no drift between your release process and your test management.

Record test results through the API

You can now submit test results into your test runs programmatically:

• Submit a result for an individual test — set its status, comment, elapsed time, assignee, and any custom result fields
• Submit results in bulk — record an entire batch in a single call
• Update an existing result when something changes

Custom result field values you submit are now returned in the API response, so your integrations always have the complete picture of what was recorded.

Validation matches the Testmo app exactly: disabled statuses are rejected, and results can’t be assigned to inactive users. The API holds the same guardrails as the UI, so automated result submission stays as clean as manual entry.

This opens the door to custom test integrations. If you run a framework or a tool that Testmo’s CLI doesn’t cover out of the box, you can now push its results straight into your manual test runs and keep everything visible in one place.

More accurate project lookup data

To support these write workflows, we’ve made the API’s lookup endpoints more complete and precise:

• The states endpoint now returns workflow states for runs, repository cases, and sessions in a single response — each one tagged with the entity it applies to — instead of run states alone. So you can resolve a valid state whether you’re working with a run, a case, or a session
• The statuses endpoint now returns only the statuses actually assigned to a project, and tells you whether each one is active or inactive — so you always submit a valid status

Plus a round of API correctness fixes

Alongside the new capabilities, this release sharpens the existing API:

• Updating a run now correctly applies the closed state
• Threaded automation runs resolve their case IDs correctly
• The Cases API date filter now includes results on the boundary timestamp
• Validation and not-found error messages are clearer and more consistent across filter parameters

Build it into your workflow

With milestone management and result recording now available, the Testmo API covers the full testing lifecycle — read and write. Whether you’re automating releases, feeding results from a custom framework, or building your own reporting layer, you can now do it through a single, consistent API.

Where this is heading: the Testmo MCP

There’s a bigger reason we’re investing so heavily in the write side of the API. These endpoints are the foundation for the official Testmo MCP — a way to connect your test repository directly to MCP-compatible AI assistants like Claude and coding tools like VS Code and Cursor. The idea is simple: your test cases already capture the edge cases, validation rules, and business logic that live nowhere else, so why not let your AI coding assistant learn from them before it writes a line of code? The same write capabilities we’ve shipped here are what let an assistant create milestones, record results, and keep your repository current as it works.

The Testmo MCP is currently in private beta, and we’re now opening it up to more teams. If you’re using AI coding assistants day to day and have a substantial test repository to draw on, we’d love to have you try it and tell us what works. You can read more about the Testmo MCP here, and apply to join the beta using this short form. Spots are limited, and beta participants get preferred access to our AI and MCP features as they roll out.