beaugunderson

i like it in theory but there’s something in me that doesn’t want emails just discarded like that… i want to be off the list. a personal failing, i’m sure. :)

ah, quarantine makes sense

i handle the overload a different way... i age everything out of my inbox one day at a time, each gets a label showing the remaining days; they start at 7 and count down. that way i only ever have 7 days of email to look at. it also makes it clear who the big offenders are... if 10% of the emails are some new marketing thing it's easy to see and unsubscribe. i do hate that there's still so much manual work involved though, i would love if marketers had to pay.

> People shouldn't be able to send emails to you without your prior approval (which you can revoke at an instant).

this sounds terrible... i've gotten so many useful random emails from strangers over the years.

the CLI is also Microsoft's preferred way to use it with agents:

> playwright-cli is best for coding agents (Claude Code, GitHub Copilot, etc.) that favor token-efficient, skill-based workflows. CLI commands avoid loading large tool schemas and verbose accessibility trees into the model context.

they do still recommend the MCP for long-running autonomous workflows (e.g. a UAT agent).

sometimes i'll share prompts but but never a whole transcript (have not had a reason to)

if i do want to share context i'll use something like "give me a prompt $coworker can share with their claude to continue this work"

totally, I see that syncing workflow come up quite a bit as well... some of this stuff is going to get sherlocked into the harness proper (probably a net benefit) though of course people are still going to want the meta-version that works with all of the different harnesses they use... so much of being effective with this style of development is molding it to your specific workflows and sanding off the rough edges with context, then sharing those wins with the larger team (another layer where I see a lot of differing approaches, like skillshare for example)

semantic search has been pretty good, it usually finds what it's looking for!

a couple of times I was certain that there was a session that contained some word but in reality it was in my personal claude.ai web account, so needed to add the import functionality there.

my favorite piece is the `corrections` command which surfaces all my frustrations/corrections in the last week for example... and I can then figure out if missing context would improve those scenarios going forward

Ah, it's an internal tool but maybe worth open-sourcing at some point! :)

re: price, definitely on the pricier side relative to our average UAT run but this one was very comprehensive due to the increased risk of issues when changing MFA... and relative to a human's time, even a cheaper QA person. I would estimate ~90 minutes of human time to figure out the needed flows and then go through the same steps on this specific change (potentially getting stuck on several pieces which the agent was able to avoid via code inspection).

> Once we trust AI to do the verification as well we'll realise the gains we feel we're getting now.

I built a UAT agent on top of claude-agent-sdk, it uses Playwright and can spin up a preview instance for PRs we open. It uses its knowledge of the code to create a test plan, runs that test plan, and takes screenshots as it does so. On a recent PR I made a change to our MFA implementation and assumed that I would need to test it myself since MFA requires setup (and a TOTP device!) but no, the agent drove Playwright to change the instance configuration to enable MFA, created a user with MFA enabled, wrote the tiniest (handful of lines) Python TOTP and used that as the device and successfully verified that MFA still worked as expected. I did not expect that ability, and I wrote it! It was also exhaustive in its probe of functionality, way more than I had planned to do, even though I had planned to do a lot given the critical nature of MFA.

Total token price for the UAT run: $5.69.

i have a claude-agent-sdk slack bot i built for work; it offers proactive interaction as well but it uses a deterministic filter and then a haiku pass for messages that pass the filter... negligible token cost.

ah, yeah, first showed up in April of 2025

untrue, I’ve deleted it many times in the last year. I don’t think this is new.

maybe I was on the wrong side of the early release but I’ve deleted this model many times in the last year. I’ve had it for at least 12 months.

opus 4.6 extended still fails.