@MHablich

Product Lead for @ChromeDevTools. Ideas my own. he/him

 Österreich
Joined November 2013
  • Tweets1.7K
  • Following273
  • Followers1.1K
125Media
Michael Hablich retweeted
API Conference@api_conferenceSep 24
🚀 API Conference #Berlin session highlights: AI agent governance, integration, APIs as product surfaces & agent interfaces — with Frederick Elewaut, @dret, Alpesh Pawar and @MHablich. 📍 Berlin 📆 November 16–20, 2026 👉 apiconference.net/berlin/pro… #APIConference
1
53
Michael Hablich retweeted
Mathias Bynens@mathiasMay 13
That’s a first: GitHub’s Advisory Database (incorrectly) flags Puppeteer as malware. github.com/github/advisory-d… We are no longer able to publish new releases. Anyone at @github or @Microsoft who can escalate this, please?

False positives w.r.t. malware in Puppeteer: GHSA-grrc-v84p-qwv3 & GHSA-rvxm-vq55-8p53 · Issue...

There seem to be two false positives that are affecting Puppeteer-related packages: @puppeteer/browsers is being flagged as malware in GHSA-grrc-v84p-qwv3. puppeteer-core is being flagged as malwar...

github.com
5
10
53
12,528
We are entering the era of dual interface products: - Rich UI for humans - Token efficient API for agents Your API is costing agents money. I'm speaking at Agentic DevDays in Stockholm on April 28 to share how we are addressing this for @ChromeDevTools .
🤖 Made with AI
68
Michael Hablich retweeted
Addy Osmani
@addyosmaniApr 12
Want to give your agent quality checks? Chrome's DevTools MCP now includes: ⚡️ Performance checks via Lighthouse 📈 Memory leak detection Skill 🦻 Accessibility debugging Skill 🎨 LCP optimization Skill and an experimental new CLI 👀
41
149
10
1,224
130,805
@MistralAI gentle ping 😺
59
Oh, that's going to be interesting (in a good way)
Peter Steinberger 🦞
@steipeteMar 28
The next version of OpenClaw is also an MCP, you can use it instead of Anthropic's message channel MCP to connect to a much wider range of message providers. (I know, this is awkward)
111
Your #AI agent isn't picking the wrong tool because it's dumb. It's picking the wrong tool because your description didn't help it pick the right one. Link: hablich.dev/articles/discove…
59
Wrote about something that's been on my mind: every browser #AI agent tool runs on CDP, which has zero auth. The security story is entirely in the layers around it. Checkout hablich.dev/articles/trust-b…
35
7. If you had to set a performance budget for agent users, what would you put in it? #WebPerf #AIagents
28
6. The mental models still hold: set a budget, measure, optimize the critical path. But the budgets change, the measurements change, and what counts as "critical" changes when the user doesn't have eyes.
1
28
5. INP measures responsiveness to human input. Agents don't care about visual feedback, but they need the page ready for the next action. Click a button, immediately fill the form that appears, page still processing? Broken state.
1
13
4. Latency: a slow page feels sluggish to a human. An agent making 200 calls turns 500ms/response into ~2 minutes of dead time. That's not a perception problem. It's a cost problem.
1
12

No more items