Ride

A Course is a saved ordered sequence of HTTP requests. A Ride is the action of executing that Course.

Store repeatable flows such as login → current user → profile update as Courses, then Ride them whenever you need to verify the flow. A collection can contain multiple Courses.

Create Courses

Use the Course selector on the Ride screen to switch the current Course, and choose New Course to add another execution path.

Each Course stores its own:

• Course name
• Ordered endpoint steps
• Stop-on-failure setting

Insert endpoints

Use the search-and-insert palette instead of checkbox selection.

Choose Insert endpoint, search by request name, method, or URL, then click a result or press Enter.

The same endpoint can appear multiple times, which makes macro-style Courses possible:

POST Login
GET Current user
PATCH Profile
GET Current user

Use the + control before a step to insert a new endpoint at that exact position.

Ride a Course

Choose Run Ride to execute every step in the selected Course from top to bottom.

A step passes when the HTTP status is below 400 and request scripts complete without errors. If any step fails, the Ride result is FAIL.

Enable Stop on failure to stop at the first failed step. Manual cancellation is reported as STOPPED.

Ride uses saved requests

Ride executes the last saved request, not the draft currently in the editor. Save with Cmd + S before running.

Configure and Ride Courses through MCP

When Free Rider MCP is enabled, an AI agent can inspect, configure, and execute Courses.

• list_courses: list Courses in a collection
• get_course: inspect ordered endpoint steps
• set_course: create or replace a Course from request IDs
• delete_course: delete a Course
• ride_course: Ride a saved Course

set_course.requestIds preserves order and may contain the same request ID more than once.

Pass values between requests

Runtime Vars stored by response extraction or After Response scripts are available to later steps.

const body = res.json()
ctx.vars.set('accessToken', body.accessToken)

Authorization: Bearer {{accessToken}}

Course-bound Mock scenarios

A Course can bind a named Mock profile through Course Mock Scenario. Riding it temporarily activates the profile on the local Mock Proxy and restores the previous selection when execution completes. The Mock server must already be running for the same collection. Concurrent scenario changes are blocked while the Course executes.

MCP set_course also accepts mockScenarioId. Use an empty string to leave the current Mock profile unchanged.