It seems like asking Claude to keep notes somehow would work better. An AGENTS file and a TODO file? An issue tracker like beads? Lots of things to try.

For a game like anchorhead, which is famous in its niche, shouldn’t Claude already know it sufficiently to just solve it right away? I would expect that its data source contained multiple discussions and walkthroughs of the game.

> By the time you get to day two, each turn costs tens of thousands of input tokens

This behavior surprised me when I started using LLMs, since it's so counterintuitive.

Why does every interaction require submitting and processing all data in the current session up until that point? Surely there must be a way for the context to be stored server-side, and referenced and augmented by each subsequent interaction. Could this data be compressed in a way to keep the most important bits, and garbage collect everything else? Could there be different compression techniques depending on the type of conversation? Similar to the domain-specific memories and episodic memory mentioned in the article. Could "snapshots" be supported, so that the user can explore branching paths in the session history? Some of this is possible by manually managing context, but it's too cumbersome.

Why are all these relatively simple engineering problems still unsolved?

Cool! I would like to see the game sessions.

Edit: they are there in the repo: https://github.com/eudoxia0/claude-plays-anchorhead/tree/mas...

Claude code, nethack, and tmux are fun to experiment with.

I’m currently letting Claude build and play its own Dwarf Fortress clone, as an installable plugin in Claude Code

https://github.com/brimtown/claude-fortress

This is a great idea and great work.

Context is intuitively important, but people rarely put themselves in the LLM's shoes.

What would be eye-opening would be to create an LLM test system that periodically sends a turn to a human instead of the model. Would you do better than the LLM? What tools would you call at that moment, given only that context and no other knowledge? The way many of these systems are constructed, I'd wager it would be difficult for a human.

The agent can't decide what is safe to delete from memory because it's a sort of bystander at that moment. Someone else made the list it received, and someone else will get the list it writes. The logic that went into why the notes exist is lost. LLMs are living the Christopher Nolan film Memento.

This is a great framework to experiment with memory architectures.

Everything the author says about memory management tracks with my intuition of how CC works, including my perception that it isn't very good at explicitly managing its own memory.

My next step in trying to get it to work well on a bigger game would be to try to build a more "intuitive" memory tool, where the textual description of a room or an item would automatically RAG previous interactions with that entity into context.

That also is closer to how human memory works -- we're instantly reminded of things via a glimpse, a sound, a smell... we don't need to (analogously) write in or search our notebook for basic info we already know about the world.

I tried something similar, but distilled to "solve this maze" as a text adventure, and while it usually solved it eventually, it almost always backtracked through fully-explored dead ends multiple times before finally getting to the end. I tried raw chat and various approaches of getting it to externalize the state (i.e. prompting it to emit the known state of the maze after each move, but not telling it exactly what to emit or how to format it), and even allowing it to emit code to execute after each turn, but it invariably would get lost at some point.

Of course it had no problem writing an optimal algorithm to solve mazes when prompted. I thought the disparity was interesting.

Note the mazes had the start and end positions inside the maze itself, so they weren't trivially solvable by the "follow wall to the left" algorithm.

This was last summer so maybe newer models would do better. I also stopped due to cost.

Having read through the entire game session, Claude plays the game admirably! For example, it finds a random tin of oily fish somewhere, and later tries (unsuccessfully) to use it to oil a rusty lock. Later it successfully solves a puzzle inside the house by thoroughly examining random furniture and picking up subtle clues about what to do, based on it.

It did so well that I can't not suspect that it used some hints or walkthroughs, but then again it did a bunch of clueless stuff too, like any player new to the game.

For one thing, this would be a great testing tool for the author of such a game. And more generally, the world of software testing is probably about to take some big leaps forward.