Kiwi for TestRail
Have the agent on a Kiwi step upload its test results to a TestRail run with the TestRail CLI and return the run link as the step's output.
- Your agent works in TestRail through its own connection. Kiwi does not call TestRail and never holds its credentials.
- A workflow skill can list TestRail among the tools it requires. When you supply your agent's available tools to a skill check and TestRail is not among them, the check reports
SK15-UNSUPPORTED-TOOL. - A project Admin can add links to TestRail, with notes, under Connections and bind them to the steps that use them. The agent reads them with
list_connections. - The agent can report the tools and plugins it ran with, including its TestRail connection, with
report_configuration. - The agent submits links to its work in TestRail as the step's outputs. People review them, and sign-off stays with the people your workflow names.
What your agent can do in TestRail
- Upload a JUnit XML report to a new TestRail test run, or to an existing run by its ID, with
trcli parse_junit. - Match automated tests to the TestRail cases your team maintains through case IDs, such as C101 in the test name or a test_id property.
- Create a run for a suite, a milestone or a chosen set of case IDs, with references to the issues it covers, using
trcli add_run. - Upload Robot Framework and Cucumber results, and import or export Gherkin feature files, with the same CLI.
- Close the run after the upload with
--close-run, or leave it open so testers can add manual results.
Set up
Connect your agent to Kiwi
Add Kiwi's hosted MCP server,
https://kiwiflow.io/mcp, to your coding agent.Connect your agent to TestRail
Install the TestRail command-line tool where your agent runs. The credentials stay with your agent.
pip install trcliExample skill
Publish instructions like these as a workflow skill that lists TestRail among its required tools.
Start from the assignment's TestRail project, suite ID and JUnit report path. Upload the report with `trcli parse_junit` and match cases by their C-IDs. Leave the run open; the assigned reviewer reads the results in TestRail. Submit the TestRail run link and the passed and failed counts as outputs. If the upload fails or cases do not match, stop after one corrected attempt and report a blocker to the step owner with the trcli error message.