No wonder agent search results have been garbage lately.

2026-10-19 17:45

No wonder agent search results have been garbage lately. Some websites are now blocking agents. Yelp wants agents to pay for a data license before it lets them in. Delta won't let an agent book a flight for you. JPMorgan and Schwab are locking accounts over "security concerns." I think websites exist to serve people, and supporting agents is just a better way to serve people. Blocking them is going the wrong direction. It's like forcing people to mail a paper letter when live chat is right there. 😮‍💨

OpenCode free models are fully gone now.

Been a long fight. Lots of open source projects have spent months playing cat and mouse with the OpenCode team. Aug: 429 rate limits. Sep 17: 403 FreeTierError on every unofficial client. Oct 17: they started flagging datacenter IPs by ASN. One proxy pool gets tagged, the entire range 403s. As of now, residential IP direct connection or the paid Go plan are the only stable ways to hit OpenCode free models.

2026-10-19

MiniMax showed its hand. Space Bunny is M3.1-Flash.

1M context, video support, $0.15/$0.60 /M. A third the cost of M3, 3x faster. 62.2 on Automation-Bench, beating DeepSeek-V4.1-Flash-1017. Official word: open weights Oct 23.

2026-10-19

Claude appeal rejected.

Won't refund the subscription either. I'm half convinced the appeal email is itself phishing, because I'm pretty sure they log the IP. DHH came to Shenzhen recently and showed how he uses Claude Code over there. That's the safest way now, honestly: run Claude Code on an overseas machine, connect through Tailscale, never use it locally. So I need to rent an overseas box with a residential IP, and register and log into Claude there.

2026-11-14

Sharing my current mobile only workflow.

Until late September it was the usual setup: parked in front of my laptop, terminal open, bossing pi around, fanning out subagents, then eyeballing diffs myself and filing the PR. Glued to the chair all day, every day. Late September I moved the whole thing to my phone, Hermes runs the show, pi takes the orders. My back is very grateful. Three weeks in, still no regrets. Stack: Hermes on flagship brains (gpt-6-astra / opus-5.5), pi on mid-tier (ds-v4.1-flash / sonnet / terra) with as few plugins as possible, the two talking over a2a. Hermes reads the issues, slices them into small tasks, spins up a worktree for each one, and dispatches a pi into it. When a pi finishes it pings Hermes, who decides: ship the PR or bounce it back for rework. I get back from a full day of being a tourist, or just wake up, and a dozen PRs are sitting there done. Pretty sweet. The catch is how black-box the whole thing is. My emotional arc has been a rerun of early vibe coding, except back then I at least read the diffs. Now even that goes to a flagship model. Boxes all the way down. If you don't have data anxiety, grok-bot and chatgpt-dot basically ship this out of the box, grok-cli and codex built in, native agent-to-agent comms. Downside: your code lives on their servers, everything's moderated, and one ban takes it all with it.

2026-11-14