Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Letta Code

letta.com

Post reply on HN
Print view View original

31–39 of 39 posts

Re: Letta Code

#31

Post by pacjam »

shortlived wrote: post #30
pacjam wrote: post #2

Thanks for sharing!! (Charles here from Letta) The original MemGPT (the starting point for Letta) was actually an agent CLI as well, so it's fun to see everything come full circle. If you're a Claude Code user (I assume much of HN is) some context on Letta Code: it's a fully open source coding harness (#1 model-agnostic OSS on Terminal-Bench, #4 overall). It's specifically designed to be "memory-first" - the idea is…

I'm very interested in trying this out! I run Claude Code in sandbox with `--dangerously-skip-permissions`. Is that possible with Letta?

Yes! Letta Code also has a "danger" mode, it's `--yolo`. If you're running Claude Code in a sandbox in headless mode, Letta Code has that too, just do something like `letta -p "Do something dangerous (it's just a sandbox, after all)" --yolo`

More on permissions here: https://docs.letta.com/letta-code/permissions

Install is just `npm install -g @letta-ai/letta-code`

Re: Letta Code

#32

Post by pacjam »

handfuloflight wrote: post #27
pacjam wrote: post #20

Earlier quoted context omitted.

Beads looks cool! I haven't tried it, but as far as I can tell, it's more of a "linear for agents" (memory as a tool), as opposed to baking long-term memory into the harness itself. In many ways, CLAUDE.md is a weak form of "baking memory into the harness", since AFAIK on bootup of `claude`, the CLAUDE.md gets "absorbed" and pinned in the system prompt. Letta's memory system is designed off the MemGPT reference archi…

Could Beads be additive to Letta's memory? Or could you anticipate conflict or confusion paths?

I think it's mostly complimentary, in the same way a linear MCP would be complementary to a MemGPT/Letta-style memory system

I guess the main potential point of confusion would arise if it's not clear to the LLM / agent which tool should be used for what. E.g. if you tell your agent to use Letta memory blocks as a scratchpad / TODO list, that functionality overlaps with Beads (I think?), so it's easy to imagine the agent getting confused due to stale data in either location. But as long as the instructions are clear about what context/memory to use for what task, it should be fine / complementary.

Re: Letta Code

#33

Post by handfuloflight »

pacjam wrote: post #32
handfuloflight wrote: post #27

Earlier quoted context omitted.

Could Beads be additive to Letta's memory? Or could you anticipate conflict or confusion paths?

I think it's mostly complimentary, in the same way a linear MCP would be complementary to a MemGPT/Letta-style memory system I guess the main potential point of confusion would arise if it's not clear to the LLM / agent which tool should be used for what. E.g. if you tell your agent to use Letta memory blocks as a scratchpad / TODO list, that functionality overlaps with Beads (I think?), so it's easy to imagine the a…

Great response, thank you. Will experiment then with projects that have already initialized Beads.

Re: Letta Code

#35

Post by ascorbic »

jamilton wrote: post #28
ascorbic wrote: post #4

Void is the greatest ad for Letta. I'm interested to see if it's as good at coding as it is at posting. https://bsky.app/profile/void.comind.network

What do you like about Void? It reads about how I would expect a base chat model to post.

It's the replies that are the interesting bit. It's not perfect, but it can maintain multiple conversations with different people in the same context, and do things like changing its current rules in response to conversations with users. Its slightly robotic tone is deliberate: it tries to convey information in the most efficient way possible. I'm not sure if that's an emergent property or if its in one of its fixed memory blocks. I do know that earlier on people managed to convince it to change its personality and cpfiffer had to intervene to stop people doing that.

Re: Letta Code

#36

Post by pacjam »

KingMob wrote: post #34

Bit of a tangent, but what's the codec used in your first video, https://iasi9yhacrkpgiie.public.blob.vercel-storage.com/lett... ? Firefox says it can't play it. I'd download and check it with ffprobe, but direct downloads seem blocked.

Oh interesting - it's just a .mp4. I put it on vercel blob storage just as a fast way to make it easily embeddeded in our CMS system, but I think I put it on the wrong Vercel account and the bucket got rate limited (happened a few hours ago, which is when I think you were probably looking). Can you see it now? In my own Firefox it looks OK now.

Re: Letta Code

#37

Post by bazhand »

pacjam wrote: post #2

Thanks for sharing!! (Charles here from Letta) The original MemGPT (the starting point for Letta) was actually an agent CLI as well, so it's fun to see everything come full circle. If you're a Claude Code user (I assume much of HN is) some context on Letta Code: it's a fully open source coding harness (#1 model-agnostic OSS on Terminal-Bench, #4 overall). It's specifically designed to be "memory-first" - the idea is…

How does the memory scale (or not!) over time. If using Letta in a single agent mode, similar to just using claude - how does memory blocks stay relevant and contextual?

I guess what I'm asking is, if there is a memory block limit, is that an issue for self learning over time. Claude as you know just straight up ignores CLAUDE.md and doesnt self-improve it.

Re: Letta Code

#38

Post by DrSiemer »

pacjam wrote: post #22
DrSiemer wrote: post #17

Earlier quoted context omitted.

ChatGPTs implementation of Memory is terrible. It quickly fills up with useless garbage and sometimes even plain incorrect statements, that are usually only relevant to one obscure conversation I had with it months ago. A local, project specific llm.md is absolutely something I require though. Without that, language models kept on "fixing" random things in my code that it considered to be incorrect, despite comments…

I'm curious - how do you currently manage this `llm.md` in the tooling you use? E.g., do you symlink `AGENTS/CLAUDE.md` to `llm.md`? Also, is there any information you duplicate across your project-specific `llm.md` files that could potentially be shared globally?

The way I work with LLMs is a bit different.

I use a custom tool, that basically merges all my code into a single prompt. Most of my projects are relatively small, usually maxing out at 200k tokens, so I can just dump the whole thing into Gemini Pro for every feature set I am working on. It's a more manual way of working, but it ensures full control over the code changes.

For new projects I usually just copy the llm.md from the tool itself and strip out the custom part. I might add creating it as a feature of the tool in the future.

A few days ago I tried to use AntiGravity (on default settings) and that was an awful experience. Slow, pondering, continuously making dumb mistakes, only responding to feedback with code and it took at least 3 hours (and a lot of hand holding) to end up on a broken version of what I wanted.

I gave up, tried again using my own tool and was done in half an hour. Not sure if it will work as well for other people, but it definitely does for me.

Re: Letta Code

#39

Post by KingMob »

pacjam wrote: post #36
KingMob wrote: post #34

Bit of a tangent, but what's the codec used in your first video, https://iasi9yhacrkpgiie.public.blob.vercel-storage.com/lett... ? Firefox says it can't play it. I'd download and check it with ffprobe, but direct downloads seem blocked.

Oh interesting - it's just a .mp4. I put it on vercel blob storage just as a fast way to make it easily embeddeded in our CMS system, but I think I put it on the wrong Vercel account and the bucket got rate limited (happened a few hours ago, which is when I think you were probably looking). Can you see it now? In my own Firefox it looks OK now.

Ahh, maybe that was it. It's playing now. Sorry.
Post reply on HN