Skip to content
Bitassert
Documentation

API Testing

Guides for authoring, running and analysing tests across web automation, API and performance — plus setup, collaboration and release notes.

Execute Tests

API testcases can be run in three ways — Cloud (queued server run), Local Agent (full testcase on your machine), and Send (single step via Local Agent). Each serves a different purpose. Use Cloud for official runs saved in Test Results; use Local Agent or Send while authoring and debugging — those outcomes are not stored in the platform.

Where to run from

LocationWhat you can do
Testcase detailsRun → Local Agent or Run → Cloud
Testcases listRun on a row (cloud)
API → Test ResultsRe-run a completed run (cloud)
Step Group editorLocal Agent only (no Cloud option)

Prerequisites

  • At least one API testcase with enabled steps (see Manual or Import Postman collection).
  • For cloud runs: a with automation_api (or create one from the run dialog).
  • For Local Agent and Send: launched and connected.

How to run a test

Use Cloud for official runs tracked in Test Results.

From Testcases list

  1. Open API > Testcases and click Run on a saved testcase row.
  2. Enter a Tag (required), select a Test Run Config (required), and optionally pick major/minor release. See .
    3. API cloud run dialog with tag and Test Run Config
  3. Click Run to queue the run. Results are saved in Test Results.

From testcase details

  1. Open a saved API testcase (Cloud needs a saved testcase ID).
  2. Click Run → Cloud.
    3. API testcase Run Cloud menu
  3. Enter a Tag (required), select a Test Run Config (required), and optionally pick major/minor release. See .
    4. API cloud run dialog with tag and Test Run Config
  4. Click Run to queue the run. Results are saved in Test Results.

From Test Results (Re-Run)

Open a completed run under API > Test Results, click Re-Run, enter tag and Test Run Config in the same dialog, then click Run to queue the run.

Local Agent execution

Debug a full testcase on your machine without queuing a cloud run.

  1. Start the Local Agent on your machine.
  2. Open the API testcase and click Run → Local Agent.
    3. API testcase Run Local Agent
  3. The UI connects over WebSocket and runs enabled steps from the current editor state (including unsaved changes).
  4. Pass/fail icons appear on each step. Results are not saved in Test Results.

From step groups

Open a , click Run, and choose Local Agent only. Inner HTTP steps run in sequence; outcomes stay in the editor.

Send (single step)

Test one HTTP request while authoring.

  1. Select a normal API step (not a Step Group step).
  2. Click Send in the request panel.
    3. API step Send through Local Agent with response panel
  3. The step runs through Local Agent. Response appears in the panel (body, headers, status, validations, extractions).
  4. After a response returns, can suggest validations and extractors for that step.

Limits

  • Send is not available for Step Group steps in the testcase view.
  • Inner steps cannot be sent individually from the testcase — use the Step Group editor for per-step Send.
  • Disabled steps cannot be sent.

Note: Before you run

  • Cloud runs cannot exceed your plan's Max execution time. See .
  • Make sure the Local Agent is launched and running on your machine.
  • Local Agent and Send results are not saved in Test Results — live feedback in the editor only. Use Cloud when you need saved run history.
  • If the agent is not connected, execution fails with a local agent error.

After execution

  • Cloud — open API → Test Results for run status, per-step response and validation details, re-run, stop, cancel, mark invalid, or delete.
  • Local Agent / Send — results stay in the testcase editor (step icons and response panel). For a saved record, run via Cloud.

Expected outcome

Cloud runs queue successfully and appear in Test Results with step-level HTTP details. Local Agent runs complete with live pass/fail in the editor. Send returns the step response for quick validation. For persisted history, tags, and team visibility, use Cloud execution.