AI testing tools
Your agent writes the code in minutes. These tools check it did the job. Four test the app: e2e by TesterArmy drives web and mobile, Reticle checks each edit inside the running app, Kerno tests the backend and Tiden checks the change against what you asked for. Two test the agent: CATT for MCP servers and APIs, Cekura for voice and chat agents. 6 tools, in the order they appear in the directory. This is not a ranking.
Grouped by job: MCP Servers & Tools · AI Agent Infrastructure
Kerno
AI runtime QA for backends: tests user flows against your real stack from inside MCP-capable coding agents.
Tiden
Intent verification for coding agents: keeps versioned specs beside your code and checks agent changes against what you asked for.
e2e by TesterArmy
Open-source (Apache-2.0) end-to-end testing framework for web, iOS and Android: describe a goal in plain language, an agent drives the app, and assertions verify the result; recorded steps replay with no model calls. From YC-backed TesterArmy.
Reticle
AI app testing that runs inside your app: it walks a flow like a user and reads network calls, app state and the console so your coding agent can verify its own work; free open-source SDK for Claude Code, Cursor and any MCP agent.
Cekura
Automated QA and observability for voice and chat AI agents: simulate conversations across personas before launch, monitor production calls for instruction-following and tool-call errors, and plug tests into CI/CD.
CATT
Consumer Agent Testing Tool: open-source (MIT) Python CLI that runs a scripted or model-driven surrogate assistant (Muse or Instinct emulation, unaffiliated) against your MCP server, MCP app, website or HTTP API, checks what the service actually did and records sealed evidence you can replay and diff in CI.
What it checks, where it runs, what it costs.
Straight from each vendor's own pages. Plans move often, so the check date is below.
| Tool | What it checks | Where it runs | License | Price |
|---|---|---|---|---|
| Kerno | Backend user flows against your real stack. | A CLI (npm i @kerno/cli) used from inside MCP-capable coding agents. CI on Pro. Self-hosting on Enterprise. | Proprietary | Free: 30 test runs a month, 1 repo, 1 user. Pro $50 per developer a month, or $37.50 billed yearly. Enterprise custom. |
| Tiden | Whether a change does what you asked: it records the request, keeps specs beside your code and maps each one to a test. | Connects to GitHub or GitLab and to Jira, Linear or Notion. Works beside Claude Code, Cursor, Copilot, Codex and Gemini CLI. | Proprietary (Qase) | Starter $49/mo: 1 product, 25 verified changes. Team $199 per product a month. 14-day trial. Annual saves 20%. |
| e2e by TesterArmy | Web, iOS and Android apps end to end. Tests mix plain-language goals with normal locators and assertions. | Your machine or CI, on Playwright browsers, iOS simulators and Android emulators. Ships an MCP server and a skill for Claude Code and Cursor. | Apache-2.0 | Free. You bring the model: an API key, a gateway, a local server or a ChatGPT or Copilot login. Verified steps replay with no model call. |
| Reticle | Each edit your coding agent makes, inside the running app: network calls, app state and console, plus file and line on React. | In your app during development, through Claude Code, Cursor or any MCP agent. React, Next.js, Vue, Svelte and Angular. The hosted platform tests previews and PRs. | SDK Apache-2.0. Server, CLI and MCP packages FSL-1.1-ALv2. | SDK free. The hosted platform has a free plan. Team and enterprise pricing on request. |
| Cekura | Voice and chat AI agents: simulated conversations before launch, then production calls for instruction-following and tool-call errors. | Hosted, with an API, an MCP server and Claude Skills. Tests plug into CI/CD. | Proprietary | Pay as you go, first user free: $0.25 per voice testing minute, $0.05 per monitored call. Startup $500/mo. Enterprise custom. |
| CATT | How an assistant uses your MCP server, MCP app, website or HTTP API, and what your service actually did. | A Python CLI on your machine or in CI. Sealed runs replay and diff across builds. | MIT | Free. Scripted runs need no model or account. |
Checked 2026-10-30. Sources: Kerno pricing, Tiden pricing, e2e by TesterArmy, Reticle, Cekura pricing, CATT on GitHub.
Pick it if.
- e2e by TesterArmyYou want end-to-end tests in your repo that mix plain-language steps with real assertions, on web and mobile.
- ReticleYour agent says it's done when it isn't, and you want it to prove each edit inside the running app.
- KernoThe risk lives in your backend, and you want flows tested against the real stack from inside the agent.
- TidenTests pass but the feature is wrong. You want each change checked against the request.
- CATTYou ship an MCP server or API and want to see what an assistant really does with it, with evidence you can diff in CI.
- CekuraYou run voice or chat agents and want simulated calls before launch and monitoring after.
Each listing.
The directory page, alternatives in the same category, and a head-to-head where we have one.
- KernoAlternatives
- TidenAlternatives
- e2e by TesterArmyAlternatives
- ReticleAlternatives
- CekuraAlternatives
- CATTAlternatives
Category pages list every tool in those categories, including ones left off this page. Built one of these? Get it featured in its category.
The 2 biggest AI stories, every morning.
One short email: the two stories that matter, and why.
You're in.
The Daily AI Brief will land in your inbox soon. If you don't see it, check Promotions or Spam and move it to your main inbox.