Skip to content
AI News HubLIVE
More
In-site rewrite1 min read

Show HN: Argus – An AI QA engineer: give it a URL and it tests your app

Summary

Argus is an AI QA engineer that takes plain English test descriptions, navigates your app like a human, and reports issues with video evidence. No selectors needed, so it withstands UI changes. It can also be called by coding agents via MCP to verify their work.

SourceHacker News AIAuthor: canergl
Show HN: Argus – An AI QA engineer: give it a URL and it tests your app
Report an error

The correction channel is not available yet. You can copy the article reference below for later.

Correction instructions
Read article

We ship most of our code with coding agents now, and testing became the problem: agentic coding is fast, but we can't test as quick as agentic coding.

That's a bottleneck.

So we built Argus.

You describe the flow in plain English:

  • "verify checkout handles an empty cart"
  • "register a new user with mail [email protected] and password Test123!, then comment on the

first post and check it appears"

  • "apply discount code SAVE20 and check the total actually drops"
  • "add two items, remove one, and check the cart badge updates"

It signs in, clicks through the app like a person, and reports where it broke, with a video. No selectors, so nothing breaks when the UI changes. Our coding agents also call it over MCP to check their own work.

We spent a considerable amount of time making it closer to a real human tester.

Feel free to create an account and try it free: https://argustest.com/signup

We're happy to hear any feedback, good or bad. If it breaks on your app, tell us and we'll dig in.

Key points and analysis

Article intelligence

EngineersBeginner

Key points

  • Argus accepts natural language test steps and executes them autonomously.
  • Operates without fragile CSS selectors, adapting to UI changes.
  • Produces video recordings of test sessions for debugging.
  • Integrates with coding agents through MCP for self-checking.

Highlights and analysis are generated automatically and may contain errors. Check the original source.