AI-powered test automation

Your AI copilot for smarter test automation.

Understand the context. Generate the tests. Automate the execution. Verify the evidence.

Bring a spec, a design, a document, notes, or code. TestNeo uses what you connect to generate tests in natural language and run UI and API checks—so your team spends less time creating tests and more time verifying what matters.

Explore TestNeo Book a Demo Start free Try playground Add to Chrome — free

Test with intent. Control your AI. Release with evidence.

The harder problem

More tests. Same blind spots.

Generating more test cases is easy. Knowing whether your critical user journeys and API behaviours are adequately tested is harder.

Coverage

Are you verifying the behaviour that matters?

A larger suite is not the same as meaningful coverage. TestNeo starts from natural-language intent and the context you connect — specs, designs, defects, and journeys — so the check is something a person can review.

Maintenance

How much of the work is still script upkeep?

Generating Playwright or API scripts is fast. Keeping them aligned as the product changes is not. Describe the check in natural language and run it, alongside the automation you already have — not instead of it.

AI cost and control

Who chooses the model, and when it runs?

Live runs, Predict, and PR risk do not need a public LLM. Generation is optional: TestNeo managed AI, your own Groq or Azure key, or a private OpenAI-compatible endpoint you make reachable.

Engineering context

Testing that uses engineering context.

Use available project context, previous failures, and change information to make testing workflows more informed and investigate issues faster. A source counts only after you connect it. TestNeo does not learn from every run on its own.

App knowledge

Figma · Swagger · PDF
Specs & designs as context

User journeys

Mixpanel · Amplitude
Hotjar · Clarity sessions

Defects

Jira · bug reports
Historical failure memory

Incidents

Sentry breadcrumbs
Production error paths

Code changes

PR diffs · impact
What changed this release

Web and API suite

Checks from the context you connected.

  1. Add missing checks
  2. Run what this change needs
  3. Update when the product changes
How coverage works

Your stack

Complement your existing automation.

TestNeo does not have to replace Jira, Playwright, CI, or the editor you already use.

Use it next to them — from Studio, MCP, the CLI, VS Code, or the Playwright AI SDK.

AI assistants

Access supported TestNeo capabilities through MCP, including Claude Desktop and Cursor.

 IDE workflows

Use the VS Code extension, or an MCP client in Cursor, Windsurf, or VS Code.

 CLI and CI/CD

Run configured checks from the TestNeo CLI and from pipelines such as GitHub Actions.

 TestNeo Studio

Manage workflows and inspect execution results, engineering context, and release evidence.
Agent SDK pip install testneo-agent-sdk · CrewAI · LangGraph · LangChain Chrome Quick Test NLP on any page → Run live → playground VS Code Extension Right-click → Generate Verification Playwright AI SDK import { test } from "@testneo/playwright-ai-sdk" TestNeo CLI $ testneo run checkout.testneo MCP AI Agent Claude Desktop · Cursor · Copilot n8n templates Import HTTP workflows → gate releases

AI cost and control

Your testing workflow. Your model choices.

These are different controls. Running a check does not call a public model. Generating tests is optional, and you choose whose model does that work.

Run

No public LLM required

Runs and release scores stay on your app and the evidence. They do not ask a public model whether to ship.

Generate (optional)

Ours or yours — same seat price

Use TestNeo’s AI, or connect your own Groq or Azure key. Skip this if you already have tests.

Private gateway

Your model, on your network

Connect a private endpoint, such as Ollama with Qwen, Mistral, or another model you host. It has to be reachable from TestNeo, not only on your laptop.

Start free AI provider FAQ

Release evidence

Know what your tests actually verified.

A passing run is not proof the release is safe. Where impact data is available, TestNeo puts results next to the change and past failures. The call still uses the checks you ran, the rules you configured, and the evidence — not a model saying the release is fine.

Without TestNeo

  • • You know tests passed or failed
  • • You don't know what changed caused which failure
  • • You run everything every time
  • • Evidence is scattered across logs and screenshots
  • • Ship decision is a gut call
  • • LLM-first tools go quiet if you turn the model off
Recommended

With TestNeo

  • ✓ Changed files map to checks when impact data is available
  • ✓ Impact analysis shows what to run and why
  • ✓ Run the suggested subset when you choose to, instead of the full suite
  • ✓ Evidence collected and unified in one release record
  • ✓ Ship decision with a clear next action and reason
  • ✓ Run, Predict, and PR risk work without a public LLM; generate is optional BYOK
See full feature comparison →

How teams use TestNeo

Same copilot. A path for each role.

QA, developers, and product start from the context they already have. Then they generate, run, and verify.

QA / Test Engineers

Write the check in plain English and see exactly what failed.

Developers

Check the change you just made before it merges.

QA Managers / Test Leads

See which checks matter for this release.

Product Managers

Turn a design or a ticket into a check you can run.

DevOps / SRE

Run the same checks in the pipeline you already use.

Engineering / Business Leaders

See whether this release is ready, and why.

Studio, Chrome, and your pipeline run the same checks.

From a client

Brochure websites, checked before they go live.

Dataworx uses TestNeo to identify visual changes and page-quality issues across static websites and contact pages.

TestNeo helped us identify visual changes and page-quality issues on our brochure websites and contact pages, giving us an additional layer of confidence before publishing updates.

Manjunath Y N, Founder, Dataworx

Works with the AI tools you already use

Connect to TestNeo through MCP and access supported testing workflows from your AI development environment.

Explore MCP Integration

Example · Claude + MCP

Your AI assistant can do more than generate tests.

Connect Claude to TestNeo Studio through MCP. Work with project context, generate and execute tests, investigate failures, and access supported release-validation workflows from your AI development environment.

Change impact

“What could be affected by this API change?”

Explore available dependency and change-impact information to identify areas that may need verification.

Test execution

“Run the relevant tests and explain the failures.”

Execute supported checks and inspect the actual execution results.

Release readiness

“What does the evidence say about release readiness?”

Review available validation results and release assessments.

Illustrative prompts. What comes back depends on the project context you connect and the actions you approve. This is not a recorded run.

Explore the MCP Integration Claude Desktop setup

Integrations

Works with your entire stack

One extension, every workflow. TestNeo plugs into where you code, how you deploy, and what you test on.

Code
Map change & generate verification
Chrome
Quick Test extension
Live
VS Code
Primary IDE
Live
Cursor
AI-first editor
Live
Windsurf
Agentic IDE
Live
Kiro
AI assistant
Live
Codespaces
Cloud IDE
Live
Gitpod
Cloud dev env
Live
push
Build & test
Automated pipelines
TestNeo CLI
Terminal / CI
Live
GitHub Actions
CI/CD
Live
Jenkins
CI/CD
Live
GitLab CI
CI/CD
Live
CircleCI
CI/CD
Live
Azure DevOps
CI/CD
Live
Jira
Tracking
Live
run
Execute
Cloud & local Chrome
TestNeo Cloud
Managed infrastructure, zero setup
Live
BrowserStack
Real devices via Chrome
Live
LambdaTest
Parallel execution at scale
Live
Local Agent
Run on your machine, behind your VPN
Live
Data stays on your machine
Works behind VPN & firewalls
Real browser on localhost
Windows, macOS, Linux

Pricing

Test automation shouldn't come with an unpredictable AI bill.

NLP web and API testing is on every plan. A run does not need a public LLM, and generation stays optional.

Bring your own model on paid plans. Use a provider key, or a model you host — such as Ollama with Qwen or Mistral — on an endpoint TestNeo can reach. Your seat price stays the same.

FreeGenerate verification
StarterBuild automation
ProScale QA
Release IntelligenceShip safely

Free

$0

14-day trial · no credit card

  • 1 project
  • 50 AI-generated test cases
  • 50 test runs / month
  • 25 PR validations / month
  • 30 MCP actions / day
  • NLP web & API generation
  • Playwright SDK spec export
  • Community support
Start free

Starter

$39/user/month

Billed monthly

Everything in Free, plus

  • 3 projects
  • 150 AI-generated test cases / month
  • 400 test runs / month
  • 50 PR validations / month
  • Figma and Swagger imports included
  • Local AI agent for runs
Get Starter
Most popular

Pro

$99/user/month

Billed monthly

Everything in Starter, plus

  • 10 projects
  • 500 AI-generated test cases / month
  • 800 test runs / month
  • 200 PR validations / month
  • Visual regression
  • Performance, accessibility, and run quality
  • Code impact, PR risk, and a release confidence score
  • CI/CD export
Get Pro
Release Intelligence

Release Intelligence

$149/user/month

Billed monthly

Everything in Pro, plus — Should we ship? Evidence from engineering history.

  • 20 projects
  • 2,000 AI-generated test cases / month
  • 2,500 test runs / month
  • 800 PR validations / month
  • Engineering Memory from Jira, incidents, and past releases
  • Release bundles with PASS / WARN / BLOCK and GitHub checks
  • Executive release dashboard
  • Priority support
Get Release Intelligence

Go deeper when you need detail

Run the check from Studio, Chrome, or an API workflow. Each page has the setup.

Chrome · New Quick Test extension NLP smoke tests from any page Automation NLP web & API Plain English → Playwright API API Risk Discovery cURL → variants → Confidence Brief Studio TestNeo Studio Generate · review · run Visual Visual regression AI Smart diffs · batch approve Performance Lighthouse & Quality Vitals · a11y · console Load k6 Load Insights Export · run · analyze Release Release intelligence Impact · confidence · ship signal Agents Agent verification Claimed → Observed → Verified MCP Claude Desktop Chat → verify → brief Memory Engineering Memory Jira · incidents → PR matches Compare How TestNeo compares Vs traditional tools

FAQ

Common questions, clear answers

Can I run load tests with the same API suites?
Yes. Export any API NLP suite as a k6 kit from Test Manager — same flows you already run functionally. Tweak VUS, DURATION, and BASE_URL locally (or in CI), run with Grafana k6 or Docker, then upload k6-out/testneo-import.json to Load Insights for p95, error rate, throughput, and pass/warn/fail verdict history. Full walkthrough: k6 Load Export & Insights docs.
What counts as one PR validation on my plan?
One PR validation = one release review workflow for a pull request or branch diff. TestNeo runs impact analysis on changed files, assigns a risk score (0–100), highlights test gaps, and produces a release brief you can open at /web/pr-validation/<workflow_id>. Triggers: MCP in Cursor/VS Code, the PR Validation page, or GitHub auto-validation. Each new PR (or first validation of that PR) uses one credit; re-opening or updating the same workflow does not. Optional test execution inside a validation uses your separate test runs allowance. Pro includes 200/month; Release Intelligence includes 800/month with full PASS/WARN/BLOCK and GitHub Check Runs.
Can you build automation for us if we don’t buy TestNeo?
Yes. TestNeo Labs engagements run in your repo and CI — the platform is optional. We use the same operator playbooks as the Academy bootcamp and cookbook, with documentation and handoff so your team can run releases after delivery.
Isn’t AI / vibe coding enough?
AI tools generate outputs quickly; they rarely deliver stable CI, clear ownership, release gates, and evidence leadership trusts. We implement that operating layer — governance, playbooks, and human judgment — not another pile of specs in CI.
What is TestNeo.ai?
TestNeo is an AI test automation copilot. You bring context, generate tests in natural language, run UI and API checks, and inspect the evidence. The same supported workflows are available in TestNeo Studio and through MCP in Claude Desktop, Cursor, and other clients. Release Intelligence is the next layer: it places those results next to the change. Regression Intelligence can use Mixpanel, Amplitude, Hotjar, Clarity, and Sentry, or an uploaded file, when you connect that source.
What is Regression Intelligence?
A suite recommender on Release Readiness: connect or upload production analytics, refresh analysis, open a ranked run pack in Test Runner, and Generate drafts from knowledge gaps (you approve creates). Guide: Regression Intelligence docs.
Is TestNeo an AI testing tool?
Yes. It generates tests in natural language, runs UI and API checks, and keeps the evidence. Release readiness is a later step: PASS, WARN, or BLOCK from the checks and rules you configured, for applications and AI agents.
We already use Claude / Cursor. Why TestNeo?
Coding agents plan and edit. TestNeo executes verification against your systems, retains evidence, matches Engineering Memory, and returns PASS / WARN / BLOCK for releases and agent runs. Bring your agent; bring your release brain (TestNeo).
We use CrewAI or LangGraph. Do we need another agent platform?
No. TestNeo is the verification and release layer for agents you already run. Install testneo-agent-sdk, ingest a thin run summary, get Claimed → Observed → Verified with PASS / WARN / BLOCK. See the AI Agent Verification guide and integration steps.
Aren’t evals / LLM-as-judge scores enough?
No. Judge scores and datasets measure writing quality. They do not prove a refund, confirmation, or memory isolation happened. TestNeo Verify is the ship decision for one run (PASS / WARN / BLOCK). Eval is oracle health across many gates — not an LLM beauty bench.
Does TestNeo auto-heal failing assertions?
Heal is suggest only. Auto-rewriting assertions can hide real product breaks. A human reviews the evidence and owns PASS / WARN / BLOCK. Generation from chat is optional support for some product paths — not the release decision.
Do I need coding skills?
No. TestNeo is designed for all users, from non-technical testers to automation engineers. Use natural language to create tests, or leverage the VS Code extension for code-first workflows.
What is NeoSense™ AI-Q?
NeoSense™ AI-Q is our AI Quality Intelligence layer. It's a context-aware assistant that lets you interact with your test data using natural language. Ask questions about test execution, get intelligent answers, and receive recommendations based on your test history.
Does it integrate with CI/CD?
Yes. TestNeo integrates with Jenkins, GitHub Actions, GitLab CI, CircleCI, Azure DevOps, and more so teams can run targeted verification, collect evidence, and enforce release gates in CI/CD.
What is adaptive execution?
Adaptive execution is TestNeo’s ability to adjust when UI or API details shift, while preserving the intent of the verification flow and recording what changed as part of the release evidence.
Can I migrate existing tests?
Yes. TestNeo can automatically convert existing automation scripts from Selenium, Cypress, and Playwright into natural language tests, making migration seamless with zero rewrite required.
Is my data secure?
TestNeo prioritises security with encryption, role-based access control, and org-scoped credentials. The Local Agent option keeps browser/API execution on your machine. For AI features, you can use TestNeo managed AI or bring your own Groq, Azure OpenAI, or a private OpenAI-compatible gateway (Ollama, vLLM, LiteLLM). Org keys are encrypted at rest. See AI providers.
Can we use our own Ollama or self-hosted LLM?
Yes. In Settings → AI Providers, choose Private / OpenAI-compatible and set your HTTPS /v1 base URL plus model names (from ollama list or your gateway). From TestNeo SaaS, the endpoint must be reachable over HTTPS (Cloudflare Tunnel, reverse proxy, ALB, or PrivateLink)—not localhost. PR validation, release readiness, and agent verification stay deterministic either way; the model is used where AI authoring and chat need it.
Does a live NLP run send steps to Groq or OpenAI?
No on the happy path. NLP execution is a real browser (Playwright) plus selector memory. A language model is a last resort for unmatched steps or leftover Studio chat — not every click. Evidence (screenshots, logs, oracles) is what the run records.
Do you send our tests to a public LLM?
Not for compile-first surfaces: live run, Predict, PR PASS / WARN / BLOCK, release gates, and agent oracles. Optional generation and Help chat use TestNeo managed AI or your org BYOK (Groq, Azure, Ollama-compatible). Semantic matching uses MiniLM embeddings on TestNeo, not your chat provider. See AI providers.
Can we use TestNeo with no language model at all?
Yes for run and release decisions. Import or write NLP yourself, execute locally or in cloud, and still get impact, memory, and gates. You skip Studio generation and leftover chat. That is compile-first — not an air-gap claim and not “no AI” (embeddings and oracles still exist).
Do we pay less if we bring our own model?
No — seat pricing is the same. You pay TestNeo for the release control plane (Studio, PR risk, gates, memory, seats). You cover your own GPU or cloud LLM costs. We don’t mark up your tokens when you bring your own provider.
Who can use TestNeo.ai?
Engineering teams across roles — developers, QA leads, SDETs, product managers, and engineering managers — use TestNeo when they need clearer release readiness, better evidence, and safer ship decisions.
How do I get started?
Install the Chrome Quick Test extension or the VS Code extension, or sign up for the web platform. Start with a 14-day free trial — no credit card required. After the trial, stay on Free or upgrade to Starter ($39), Pro ($99), or Release Intelligence ($149) per user each month.
Do I need to replace Playwright?
No. You extend it. Generation and Playwright-oriented workflows are part of application verification. The product outcome is release confidence—not script volume.
Does it run locally?
Yes, by default.
Can I use only Playwright for some steps?
Yes, fully hybrid.
Can I try without signup?
Yes. Use the live playground — starts on Web Automation (NLP), 10 guest runs, no account required.
When do I need API key/project?
For dashboard publish and team usage.

Featured on

Launched on Launch Streak Live · Launch Streak Featured on LemonLaunch Live · LemonLaunch Featured on neeed.directory Live · neeed.directory Featured on Fast Launch Live · Fast Launch Featured on SubmitMySaas Live · SubmitMySaas Fazier badge Live · Fazier Featured on DeepLaunch.io Live · DeepLaunch.io Featured on SumoDir Live · SumoDir TestNeo.ai on Nick Launches Live · Nick Launches Featured on Findly.tools Live · Findly.tools Good AI Tools Live · Good AI Tools Featured on Twelve Tools Live · Twelve Tools Million Dot Homepage Live · Million Dot Homepage We Like Tools Live · We Like Tools Tools Under Radar Live · Tools Under Radar
Chrome Web Store Live
Live · Quick Test extension

Bring us one real testing problem.

A journey that is hard to keep in scripts, an API change you are unsure about, or a release call that still depends on a green count. Start in the product, or walk through it on a call.

Start free Book a Demo

Get in touch

Let's talk about your next release

Or reach out directly

support@testneo.ai Schedule a call