Read more in our changelog: argos-ci.com/changelog/
10.03.2026 15:38
👍 0
🔁 0
💬 0
📌 0
Flaky visual tests slowly destroy trust in a test suite.
You review the same diff again and again. Nothing changed.
We solved it in Argos:
If the same screenshot diff appears repeatedly across builds, Argos automatically ignores it.
Visual tests that start to heal themselves.
Came across a solid set of plug-and-play React components, and it even includes an MCP ✨
A table displays three months of performance data representing the status of GitHub Actions.
Graph representing the status of GitHub Actions over the last 90 days.
GitHub is the most unreliable service I have ever seen. Their SLA is crashing down so quickly 😱
• Jan: < 99.9% = down 10 min per week
• Feb: < 99% = down 1h 40m per week
• March: < 98% = down 3h 20m per week
• April: ?
A cartoon depiction showing cloud servers, PostgreSQL icons, and an AWS logo, symbolizing database migration and cloud infrastructure.
One of the biggest engineering projects at Argos recently was migrating our 300GB PostgreSQL database off Heroku.
We accomplished this by building an EC2 bridge and replicating into RDS with two maintenance windows of 1 minute each.
I wrote a deep dive on the full migration 👇
Preview: nodejs-api-docs-tooling.vercel.app/
PRs: github.com/nodejs/node..., github.com/nodejs/node...
Node.js is redesigning its docs site. Here’s the current version vs. the new one 👀
Login interface featuring fields for username and password, alongside a SAML security icon and a green checkmark, indicating approval.
SAML SSO is now available on Argos Enterprise 🔐
If you’re a larger team, identity management isn’t a “nice to have.” It’s table stakes.
You can now:
• Authenticate via your IdP
• Auto-provision users on first login
• Enforce SAML across your team
We shipped a new core algorithm in Argos from an AI agent draft → Rust → production.
Used for visual diff fingerprinting, now powering key features.
~3× faster than the TypeScript version.
Breakdown + code + benchmarks:
argos-ci.com/blog/ai-gen...
Read Argos changelog to learn more about it
argos-ci.com/changelog/
AI simplifies choosing relevant tests, making full suite runs on every branch unnecessary.
Introducing Subset builds in Argos: run partial tests while maintaining clear visual feedback. Ignored screenshots streamline reviews, and full runs on main still set baselines.
We just shipped a new thing: the Tests Dashboard. It gives a project-wide view of all tests, ranked by flakiness.
No more intuition or Slack archaeology to decide what to fix first.
If flaky tests slow down your reviews or CI, this makes the problem visible in seconds.
As co-founder of Argos, this is exactly why we built it.
Pivot chose Argos + Playwright from day one.
Screenshots over brittle assertions. Visual baselines as the contract.
Stable tests, regressions caught early, faster confidence to ship.
Story 👇
argos-ci.com/customers/p...
Just published a customer story with Finviz.
For them, a one-pixel chart shift is a functional bug.
They use Argos with Playwright for pixel-level diffs in CI and PRs, with zero extra test maintenance.
argos-ci.com/customers/f...
Proud to share this one.
MUI runs visual regression testing at massive scale with Argos:
thousands of screenshots per build, millions per month, across OSS repos.
Fast PR feedback, low noise, real regressions only.
Full story 👇
argos-ci.com/customers/mui
Argos now runs natively inside the merge queue, with correct merge bases and required visual checks, no extra setup.
argos-ci.com/changelog/2...
To celebrate the start of the new year, we redesigned the Argos website ✨🥂
Featuring 3 core features:
✦ Visual Testing
✦ Flaky Management
✦ Test Debugging
Excited to see the Argos vision taking shape step by step 🙃
const obj = { a: 1, b: 2 }; const count = Object.keys(obj).length;
ECMAScript excitement 😉
Congrats to @bridgear.bsky.social @jordan.har.band on advancing Object.keysLength to Stage 2 at TC39 today 🎉
let count = Object.keysLength(o)
It counts the string-keyed own properties on an object. This avoids the cost of the intermediate array in Object.keys(o).length
We just shipped ARIA Snapshots in @argos-ci.com for @playwright.dev 🎉
Now you can capture accessibility trees alongside screenshots, perfect to spot text changes and catch accessibility regressions.
Hey Filip, thanks! Argos has always been open source 😉
Argos is getting more enterprise-ready 💜
Get notifications where you need them!
Welcome to a new era of documentation.
Starting today, you can create tailored docs experiences for every user, with an AI Assistant that can answer with context from external sources via MCP.
Create relevant, targeted docs for your users. Find out more: www.gitbook.com/blog/new-ada...
Flaky test detection is the first step in bringing full flaky test management to Argos. Visual testing is inherently flaky, but it doesn’t have to stay that way. Our goal is to give all teams the tools they need to understand, manage, and eliminate test flakiness.
We just released Slack notifications at @argos-ci.com 🎉
Real-time build updates, configurable rules, channel targeting. Already making our own workflow smoother!
Some thoughts on how to write more stable Playwright tests, starting with the locators you choose. This short guide covers what works in practice and how to avoid common pitfalls.
argos-ci.com/blog/playwr...
