Keep Fable, but have it orchestrate other agents running different models (could be Opus, Grok or Kimi K3).

This is my current workflow: https://x.com/sridca/status/2093354491965735091

I even let Fable choose the appropriate model for the task.

Glad someone noticed this.

And that is the crux of the whole issue here, as explained in https://sipping-delicious-tea.pages.dev/

> you can get access to these models with unmetered tokens with SuperGrok [$USD 300/year]

Could you support this statement with an official reference?

Probably so they thought it would circumvent insults and other wily rhetoric from those emotionally sensitive on this subject ... and yet, going by your own use of "idiot" on a website enabling otherwise intelligent discussions, it would seem that was futile consideration on OP's part.

I created https://agency.srid.ca/ to achieve some of this but from within a single agent CLI session

Here's a recent PR created end-to-end using `/do` workflow of agency: https://github.com/srid/emanote/pull/719

I just used this to summarize HN posts in last 24 hours, including AI summaries.

This PR was created by the Claude Code Routine:

https://github.com/srid/claude-dump/pull/5

The original prompt: https://i.imgur.com/mWmkw5e.png

I started working on an "anti-habituation" motivational reminder app:

https://appreciate.srid.ca/

https://github.com/srid/Appreciate

(Note: Only the macOS version is tested extensively at the moment)

If you are using Nix, there's also https://github.com/srid/sandnix that works on Linux (landrun) and macOS (sandbox-exec).

> not too much of my post history accumulates in one account

I'm curious to hear what benefits you think can be gained from avoiding this.

I do the same. It simply means theres less accidental leakage / self-doxing that could be pieced together if you (or llm) read every comment on the account.

Suggestion: Pick a long term account, dump the comments, and see what an llm could figure out about the target

I do it sometimes just to restrict my own pride in the account. I get a buzz from upvotes and that upsets me on a deeper level.

Same, but also for the opposite reason: a new account gives me a chance to do better. If I post lame comments, I accept the lameness of the posts attached to a particular user name and the hesitation I feel to post more lame comments decreases. With a fresh identity, I am more likely to avoid lame posting sort of like how you avoid going out in the mud in brand new sneakers. A sort of repentance; being born again in the digital realm.

You can build quite an extensive profile of someone given enough post history. More post history means more details. Especially nowadays with LLMs it's trivial. This can lead to all sorts of issues. One is people I know in real life being able to identify me. Another is that through various means my account may be linked to my personal identity (e.g. through matching usernames or emails across platforms) and oppressive regimes (now or in the feature) may use my post history to take action against me.

Yea, I think it is worth waiting for M6 just for OLED alone.

OLED sounds great, but I am worried about burn in. MacBook screens are a bit more static and on longer than iPads and iPhones.

I use a massive OLED monitor as my workhorse and I’d say money and expectations are better spent on established OLED manufacturers and a large screen vs. a laptop screen. Based on the common job roles HN users have, a large OLED main monitor will probably offer more value than the laptop screen that will probably spend most of its time as a side monitor or just turned off while connected to large monitors. The HDMI 2.1 and other display output gains bring more benefits with pixel output and framerate increases. Just my two cents.