Overview

AI-powered backend testing for RESTful APIs. Discover endpoints, generate plans, ship endpoint and integration tests, watch them run, and auto-clean up the data they touched.

API Testing in TestSprite is a journey from “here’s my API” to “here’s a regression suite I trust”. Each phase has its own dedicated docs page; this page is the map.

Key Features

Feature	Description
Fast Setup	Get started with minimal documentation — no extensive prompts or full codebases needed
Endpoint + Integration Tests	Per-endpoint coverage and multi-step chains where one test’s response feeds the next
Dynamic Variables	Capture values from one response, inject into the next request — TestSprite wires the dependencies for you
Auto Cleanup	After every run, TestSprite deletes the records your tests created so your test workspace stays clean
Data Flow Visualization	A grouped, navigable view of every HTTP call TestSprite made — request, response, timing, dependencies
Auto-Auth (Pro)	Configure your login flow once; we fetch a fresh token before every run
Smart Reports	Detailed outputs with error types, root causes, and recommended fixes
Natural Language Refinement	Adjust any test in natural language — no code required

The API Testing Journey

Each step gets its own page. Click through in order if you’re new — or skip ahead to the area that’s currently giving you trouble.

API Discovery

How TestSprite finds your endpoints from PRD, code, and live probes

Plan Generation & Editing

Review the plan, edit cases, add scenarios in natural language

Endpoint Tests

Per-endpoint test generation with verification before tests land in your suite

Integration Tests

Multi-step chains where one step’s output feeds the next

What Lives Where

The Web Portal sidebar for an API project organizes pages by what you’re trying to do, not by chronology. Once a project is set up, you’ll mostly bounce between these:

Sidebar location	What you see	Detail page
Overview → Use Case Flow	A diagram of your features and the endpoints touched by each	(covered as a section in API Discovery)
Overview → Data Flow	Every HTTP call from the most recent run, grouped by endpoint, with producer/consumer wiring	Data Flow
Overview → Endpoint Tests	The flat list of per-endpoint tests, status chips, last-run results	Endpoint Tests
Workflows → Integration Tests	The list of multi-step chains that exercise sequences of endpoints	Integration Tests
Workflows → Cleanup	Auto-generated DELETE chains that wipe records the tests created	Auto Cleanup
Data → Dynamic Variables	The producer-consumer graph — what each test produces and what depends on it	Dynamic Variables
Data → Authentication	Auto-Auth config + Static Credentials per API family	Auto-Auth (Pro)
Results → individual test detail pages	Per-test request/response, error trace, refine + rerun controls	Test Detail

The sidebar’s order is the recommended order for first-time setup. Walk top-to-bottom on a new project: see the use-case overview, look at the endpoint tests TestSprite generated, peek at the integration chains it identified, check what values flow between tests under Dynamic Variables, configure auth, then run.

Quick Start

Quickstart

Walk through your first API project end-to-end — from project creation to a passing HTTP suite — in about 10 minutes.

What a Plan Covers

Generated plans group test cases by category — TestSprite picks categories that make sense for your API based on what was discovered. Common ones you’ll see:

• Functional / happy path — does each endpoint behave correctly with valid inputs?
• Schema and response validation — does the response shape match what’s expected?
• Authorization & authentication — are protected routes actually protected? token handling correct?
• Error handling & edge cases — malformed inputs, missing fields, boundary values, type mismatches
• Boundary / load — what happens at high request rates or with oversized payloads
• Security probes — privilege escalation, signature tampering, common API attack patterns

Not every category applies to every API. Categories shown on your plan reflect what TestSprite considered relevant given your endpoints, docs, and PRD.

Best practice on first run: keep all categories selected. Once you have a baseline, you can prune categories that don’t apply (e.g., drop security probes for a strictly-internal service).

Refining Tests in Natural Language

Every test row has a chat affordance. You can say things like:

Refine

Test POST /orders with invalid parameters and expect a 400 error code.

Add a test for the rate-limit edge — burst 100 calls in 1 second, expect at least one 429.

For the GET /users/{id} test, drop the assertion on `created_at` — it's a timestamp and flaps.

TestSprite interprets the request, regenerates the affected test code, and re-runs. See Refining Tests for what the chat can and can’t do, and the alternatives when chat isn’t the right tool.

Frontend’s Sibling Page

If you’re testing a website (UI / browser flows) rather than an HTTP API, the frontend equivalent of this page is UI Testing — Overview. Most of the wizard, refinement, and rerun concepts are mirrored — what changes is the underlying execution model (HTTP calls vs. browser actions).

Where to Go Next

API Discovery

Walk through the wizard’s first phase — endpoint extraction

Auto-Auth (Pro)

Stop pasting expiring tokens before every run

Data Flow

Visualize every HTTP call from the most recent run

Subscription Plans

Free includes API testing — paid unlocks Auto-Auth and more