🆕 Excited to announce the Cognito book I've been working on for the past several months. Learn how to use one of the most powerful, scalable, and affordable user authentication services around. cognitobook.com

Fable will be back, likely, sooner than you think. It will be ok Go outside and hug your Opus

"Claude please revert my personality until Uncle Sam give me back Fable"

Fable seems very smart. But one thing I'm noticing is its just going to do work without me saying that ok. This includes jumping into work I haven't okayed yet and in some cases YOLOing deploys without my say so... Shoulda been named EagerBeaver Be careful out there and watch your Fable sessions 👀

Spot on

If you aren't ready to update your entire personality around the latest model drops.... YAGMI

1. npm install -g npq 2. alias npm=npq 3. 🎉 if you follow me and don't know what npq is... github.com/lirantal/npq

The sooner you experience "AI ego death", the better off you will be

Lolz this is NEXT LEVEL marketing. I love it

Does anyone have a markdown to slack message converter tool? Need to copy/paste markdown into slack and have it format nicely in there...

I went HAM on git worktrees when I learned they were a thing like 6 months ago but slowly drifted back to single branch flows. It's just way easier to manage and far less repeat/conflicting work. But how do you make sure multiple agents don't collide? The flow is simple: You need proper planning. Measure twice cut once has never been more true. After the plan is created, I prompt the model if it's 100% clear on what to do, and if not to surface non obvious questions edge cases we haven't yet covered. Now with a SOLID plan, I turn it into a directed graph of tasks where they know the clear order of operations. The task graph is then fed into a swarm of tmux sessions running [insert your favorite harness here] by a top level agent, lets call it the "ring leader" The agent sessions collaborate via file reservations and their own agent mail messaging system. This ensures no edits are conflicting and duplicate work is not done. The top level ring leader agent just sits there on a cron, checking in to the sessions every ~5 minutes to nudge them to pickup unblocked tasks from the graph if there are any until all the work is done. Then when its all done, I prompt for a summary of work completed and a QA plan for me or other agent to verify. This is like the "ralph loop" but on galactic steroids. It gets work done would have taken several weeks a year ago to get done in a ~2 hour session. If this all sounds too hard to manage, I thought so too. But then I stumbled across what @doodlestein was doing thru one of @johnlindquist's AI workshops. Then Jeffrey released his skills jeffreys-skills.md and holy cow everything I just mentioned above unlocked.

We need to put inlining all your dependencies at the top of the Security best practices list

Someone needs to tell Bernie about the Doctor Dolittle unlock That might change his tune Or maybe he'll want an additional 25% equity stake for squirrels Time will tell

Liquid templating engine from @ShopifyEng is legit Every time I get to work with it, its a treat

Agent proxies are crucial component of this brave new world we are entering Fine grained scoped API keys will become table stakes for tools that want to protect users and not be the next "whoopsie doodles" viral hack vector The shim in-between the agent creds is paramount.

Mantine 9.3 is out! Splitter component & use-splitter hook, Menu upgrades (search, checkbox/radio items, context menu, type-ahead), Popover context menu, Combobox floatingHeight, Pagination responsive layout, CodeHighlight line numbers and more mantine.dev/changelog/9-3-0/