Agentic end-to-end testing for developers, from code changes through CI and IDE workflows.
TestSprite is an AI-assisted testing platform that explores a running application, creates end-to-end tests, executes them, and reports failures with supporting evidence. It checks actual frontend and backend behavior, with tests running in remote cloud environments. Teams connect a deployment that TestSprite can reach; TestSprite does not build or deploy the application. The same testing workflow is available through its open-source CLI, an IDE connection using MCP, CI, or the browser dashboard. Passing tests are retained to build regression coverage over time.
TestSprite reads the application and can use a PRD or API documentation to understand expected behavior, then creates frontend and backend tests without requiring users to write test scripts or maintain selectors.
Agents interact with a real browser or live API to check user flows and backend behavior against the running app, rather than validating only mocked responses.
A failure report can include the failing step and neighboring steps, screenshots, DOM snapshots, test source, a root-cause hypothesis, and a recommended fix.
Passing tests remain in the project suite and can be rerun as the app changes. Paid plans include auto-healing reruns to adapt tests when the UI changes.
Run tests from the open-source CLI, connect an IDE or coding agent through MCP, gate a pipeline using structured JSON and documented exit codes, or use the browser dashboard.
The CLI can retrieve a failure bundle containing structured results and artifacts such as session video, reproduction steps, screenshots, and DOM snapshots.
Run a suite against a pull request deployment and use the structured result to gate a merge when required end-to-end checks fail.
After an agent implements a feature or fix, use the CLI verification workflow or IDE connection to test the resulting behavior in the application.
Review the failed step, screenshots, DOM state, and suggested cause to reproduce a broken flow and guide a fix, then rerun the same test.
Connect a staging URL and let TestSprite explore the product to create an initial suite, then keep passing tests as regression checks.
Use scheduled test runs and monitoring, available on listed paid plans, to check an application beyond an individual code-change workflow.
Use Node.js 20.19+ in the 20.x series, 22.13+ in the 22.x series, or version 24 or newer. A TestSprite account and API key are required. Install with `npm i -g @testsprite/testsprite-cli`, then run `testsprite setup`; `npx @testsprite/testsprite-cli` is available for one-off use.
TestSprite generates end-to-end tests from application context without requiring you to write scripts or hand-maintain selectors. Paid plans include auto-healing reruns for UI changes.
Yes. The CLI emits structured JSON with documented exit codes, and the GitHub App can test a pull request’s preview when your CI signals it is ready. To block merging after a failed run, configure the TestSprite status check as a required check. The deployment must be reachable from TestSprite’s cloud.
Failure details can include the failing action and surrounding steps, screenshots, DOM snapshots, test source, a root-cause hypothesis, and a recommended fix. The CLI can pull the run artifacts into the repository.
The monthly options include Free with 150 credits, Starter with 400 credits at $19/month after a free first month, Standard with 800 credits at $39/month, and Pro with 1,600 credits at $69/month. Free and Starter include one environment; Standard includes three; Pro includes unlimited environments. Run history lasts 30 days on Free/Starter, 90 days on Standard, and is unlimited on Pro. Enterprise starts at ten seats with customized limits. Yearly billing is a separate option.
トラフィックデータは参考情報としてご利用ください。
www.promptfoo.dev
Promptfoo is an AI security and testing platform for evaluating LLM applications, agents, models, and workflows. It helps developers and security teams find vulnerabilities, validate guardrails, map findings to security frameworks, and track remediation through development and deployment.
www.mcpjam.com
MCPJam is a testing and evaluation platform for MCP servers. It helps developers inspect servers locally, run user and model-based tests, and add behavior checks to CI/CD workflows.
fetchsandbox.com
FetchSandboxは、AIコーディング環境内でAPI連携、webhook、状態を保持するワークフローをテストできる開発者向けツールです。実際のAPIキーや本番環境の利用枠を使わずに、開発者やエージェントのワークフローで連携を検証できます。
text2test.ai
Text2Testは、デザイン、コード、チケット、仕様書から構造化されたテストケースを生成し、Webテスト自動化ワークフローでテスト計画として実行できるAIネイティブなソフトウェアテストプラットフォームです。
qagent.in
QAgent is an AI agent testing and quality assurance platform for developers and agile teams. It connects to an agent through a webhook or endpoint, runs automated test cases, and evaluates responses for groundedness, policy adherence, prompt compliance, and related quality dimensions.
deepeval.com
DeepEval is an open-source LLM evaluation framework for testing and benchmarking AI applications. It helps developers run pytest-native evaluations, score outputs and agent traces, and iterate on systems across text, image, audio, and voice workflows.
