Join the discussion
Write your take first — we'll ask for email only when you're ready to publish.
- Hacker News
- I love these. Coding agents aren't very difficult to build, it's a TUI + tools + getting a nice agent loop working. The hardest part seems to be supporting all of the different providers and model quirks. What is interesting is seeing the experimentation: some provide tons of tools, others provide a single python interpreter and have the agent use tools via sandboxed python scripts, others use minimal tools and lean on bash. Personally I want a harness that gives a ton of control to the user to let them steer the LLM, less agent and more augmentation. Maybe I'll have to build it myself. If anyone has ideas, let me know.by slopinthebag
- Pi.dev is pretty good in giving tons of control to the use and has extensions that you can easily build.
Although people are complaining about its RAM usage in this thread, I haven't bothered to check how much RAM it uses.
by afzalive - I'm working on one right now where nearly everything can be expressed as a combination of workflows. There will be some built-in agent types out of the box but all the Lego pieces are there if you want to put together something different.by inhumantsar
- Given agent harnesses affect so much of the performance of models, it would be great to see some kind of benchmark on how this tool performs compared to claude/codex/opencode/pi etc.by sergiotapia
- Hi! While I didn't try any agent benchmark, I already though of this possible issue, and I tried to approach it on two different levels:
1. The tools that are given to the agent are almost the same to the one defined in Opencode, except for Skills and Subagents (both features not implemented in zerostack)
2. Zerostack is prompt-based, so that it ships with a set of .md files, stored in ~/.config/zerostack/prompt, and that can be selected from the TUI in order to activate different 'agents': as you can see from the README, it is designed to contain the most important feautres of superpower + Claude's front-end design + git worktree support and Ralph Wiggum loops (both as integrated features)
by gidellav - We don’t trust llm execution- so we add user approvals. But task decomposition calls for co-recursion between code and prompts. This means that the approvals should be evocable at any depth. I think we need some kind of protocol for that (à la the Cubes OS protocols for cut and paste between vms).
Maybe a workaround could be to use bubblewrap of the scripts ther recursively call the llm (and run the agent in yolo inside the wrap).
by zbyforgotp - zerostack contains --sandbox flags that forces bwrap usage on all shell tool usageby gidellav
- Too late for fixing it - but of course I meant https://www.qubes-os.org/by zbyforgotp
- Transactional recursive agents ?
Nothing is committed until the final top-level transaction is accepted.
by agumonkey - Currently, having LLM feeding on its own output repeatedly is the fastest way to get it hallucinate.by hashmal
- Well, or not spawn any external commands, and actually have tools made of code written by someone who thought about what the agents at each level should be limited to doing.by frabcus
- The codebase was small enough that I handed it over to DeepSeek v4 Flash in Pi to skim through for any risky business, and I didn't find anything concerning. Nice work.
- > I handed it over to DeepSeek v4 Flash in Pi to skim through for any risky business
Doesn't prompt injection make that a rather flimsy investigation?
by kadoban - Thanks! Funny enough, a good chunk of the coding was done by Deepseek v4 Flash, while I hand-wrote a couple of the TUI logic, as deepseek kept failing on certain cursor-moving logic, and I fully managed the memory optimization process (as you can read on another comment I left, it both a set of compiler optimizations and usage of certain Rust crates in order to leverage more efficient data structures).by gidellav
- Since the OP stated they used DeepSeek V4 Flash for generating a lot of the code, I decided to check whether there were any outdated dependencies. In my experience, with Rust projects, if you do not instruct models (even Claude 4.7 Opus) to use `cargo add` instead of manually editing the Cargo.toml, you will almost certainly get out-of-date dependencies added to your project.
Manually checking the dependencies used by this project, I was pleased to see they are all the latest version. That doesn't mean there are no issues lurking in transitive dependencies, of course.
As for getting an LLM to review the code, I think we can get all opinionated very fast. For instance, when I was eyeballing the code, some of the enum methods converting to/from strings made me think "this could've been a single #[derive] with strum." That would make the code in provider.rs a lot more concise, at the cost of importing one crate (with no dependencies!)
Lastly, for fun, I decided to get DeepSeek V4 Pro (with Max thinking) to "audit" the codebase. The output mentioned no obvious signs of hidden telemetry, but it did note that the project sets the panic handler to "abort", which I have strong opinions on... Presumably the OP wanted to avoid linking against libunwind to save a few kilobytes of binary size, but now you have a binary that immediately aborts and doesn't give the user a stacktrace of what just crashed. I would rather have a ~50 KiB larger binary if it means getting useful debug info during a panic. Additionally, if there are async tasks that panic, they can't be recovered to display a generic error message; instead the whole process just aborts.
by koito17 - This is nice! I tried it for a bit and it was indeed quite fast. Are you looking for contributors, or are you building this as a personal tool? I ran into some issues when attempting to use different models, though: gpt-5.5 on Azure doesn't work, even with the OpenAI compatible endpoint, because "max_tokens" has been replaced with "max_completion_tokens". And it doesn't appear possible to pass through custom headers, so I wasn't able to specify reasoning_effort for deepseek models.by wkcheng
- Yes, I am open for PRs.
What you showed is a clear bug in my codebase, if you can, open a Github issue with each of your bugs.
Thanks!
by gidellav - Funny this comes out today. I was just about to start to write one in rust. It's amazing having opencode slowly leak memory and end up becoming 6gbs on a large project and then get slower and slower.
Will check this out! Seems cool!
- Yes! This project derived from an OOM killer activation that happened on my old laptop beacuse i had more than 2 opencode instances open together with Firefox...by gidellav
- I had Claude Code build me one of these as well, though I added Dirac's line hashing for edits etc. Also used Rust, and I had this idea that I should use plugins so it can self-edit by implementing in hooks but in the end, I just have it create exhaust information about improvements into a separate file and just update the source code and recompile. The source code is in a fixed place so it can just rewrite and build the agent itself. I use it with DeepSeek 4 Flash running on 2x RTX 6000 Pros which I get some 138 tok/s on.
To be honest, I just plagiarized Pi, Dirac, OpenCode. Any new tricks in this one that I can steal?
by arjie - Is it public on github?by teo-mateo
- Some interesting features I add on top of being lightweight are the prompts library, Git worktrees integration and Ralph Wiggum loops integrations.by gidellav
- Creator of Dirac here. Glad to see it mentioned and even more glad that you found it useful.
I am currently in deep refactor mode to introduce modular tooling to Dirac since the concept of 'fixed' set of tools is starting to feel antiquated, adding tools on demand would be super convenient and a likely replacement for MCP (I understand not all use-cases of it)
- Take a look at OpenAI blogs about codex: https://openai.com/index/unrolling-the-codex-agent-loop/ https://openai.com/index/harness-engineering/ https://openai.com/index/unlocking-the-codex-harness/by joshka
- Are agent harnesses the new web framework?
Everyone wants to write one, building a new one is easy to start with, but tough to get to “prod ready” and the landscape is littered with failed attempts?
Certainly feels like it.
This is really good though; works well and at least has a clearly articulated raison d'être.
- "RAM footprint: ~8MB on an empty session, ~12MB when working"
I like this, Claude Code is using multiple gigabytes, which is really annoying on lowend laptops
by throwa356262 - Isn't that because of the context window size?by marknutter
- I have 29 Claude Codes open, using 6.3 GiB RSS totalby rane
- Are you sure you don't have an LSP plugin or something running?by esperent
- Yes. Just this fact is going to make a lot of people try it out.by tecoholic