Advertisement
Advertisement
⚡ Community Insights
Discussion Sentiment
71% Positive
Analyzed from 2720 words in the discussion.
Trending Topics
#agent#harness#sandbox#code#api#don#model#llm#agents#loop
Discussion Sentiment
Analyzed from 2720 words in the discussion.
Trending Topics
Discussion (47 Comments)Read Original on HackerNews
A lot of this post presents false dichotomies. It assumes the existence of a sandbox that is by definition ephemeral or "cattle-like". Why? There are reasons to do that and reasons not to do that. You can have a durable computer with a network identity and full connectivity, and you can have that computer spin down and stop billing when not in use.
There are a zillion different shapes for addressing these problems, and I'm twitchy because I think people are super path-dependent right now, and it's causing them to miss a lot of valuable options.
[1]: https://fly.io/blog/tokenized-tokens/ (I work at Fly.io but the thing this post talks about is open source).
This is an interesting and novel field, so I’m not pretending I know the answers, but this is what worked for us :)
At the end of the day, and oversimplifying things: why would I want to spawn a for loop that calls an API (LLM) into its own dedicated sandbox/computer?
When the model wants to run a command, it’ll tell you so. Doesn’t need to be a local exec, you can run it anywhere, the model won’t know the difference.
The agent loop itself doesn’t need sandboxing. In many cases, most tool calls don’t require sandboxing either. For the tools that do require a computer, you can route those requests there when needed, rather than running the whole software in that sandbox.
To me running the agent loop in the sandbox itself feels like “you should run your API in your DB container because it’ll talk to it at some point”.
I've heard many claims that because LLMs are tuned to specific harnesses, we should expect worse performance with novel architectures. That seems to make people reluctant to try to put effort into inventing them.
Manus rebuilt its harness five times in six months. The model stayed the same, but the architecture changed five times.
LangChain re-architected Deep Research four times in one year.
Anthropic also ripped out Claude Code’s agent harness whenever the model improved.
Ever since Mitchell Hashimoto mentioned the harness in February, people have been trying to claim that concept. Eventually, someone will probably sell a book called Harness Engineering. I will buy it, of course. Then I will write a blog post about it that nobody reads, with a link that will be buried under ShowDead as soon as I submit it to HN.
And by that point, IT companies will start asking:
“You’re a new grad, right? You know harness engineering, don’t you?”
In my opinion, the main driver here is how fast models have evolved in the past 12 months. It makes the architecture of everything around them obsolete, very fast.
We went from using models as a building block, wrapping them in heavy workflow code, to now models being smart enough to drive their own workflows and planning.
What. The idea is as old as anyone can remember, and wrt. LLMs, it was known to be important since at least as early as ChatGPT being first released.
But I think the term started being used closer to its current meaning around this point:
https://www.softwareimprovementgroup.com/blog/what-is-harnes...
In a way, the sequence was something like:
prompt engineering(23~4) -> context engineering(25) ->harness engineering(26)
At first, it was mostly understood as a correction or extension of prompt engineering. But the idea of “harness” as the layer that corrects, constrains, and operationalizes agents seems to have emerged much more clearly around 2026.
So yes, there is definitely some terminological confusion in the early phase. That is normal. New technical fields often begin with several competing names for almost the same layer, and only later does one term become stable.
Having said that, some components need to live outside the sandbox (otherwise, who creates the sandbox?). Longer term, I see it as a dedicated security layer, not part of the harness. This probably has yet to emerge fully but it's more like a hypervisor type layer that sits outside of everything and authorises access based on context, human user, etc and can apply policy including mediate the human intervention for decision points when needed.
I don’t get it. Calling an API requires a sandbox in most cases. The others could be abused in service of an un-sandboxed agent with API access.
If the harness is outside the sandbox then it’s just an ambiguous and confusing security model and boundary.
I'm not following why this would this be the case? The purpose of calling the API is to get data or effect a state transition on some remote service, but I don't follow why the originating machine matters.
Or is your objection about auth?
I think the confusion is that “agent” is used for two very different things:
- building an agent
- an “agent” product/runtime (Claude Code, etc)
In the first case, the model never executes anything. It just outputs something like “call this API”. Your code is the one doing it, with whatever validation you want. There’s no need for a sandbox there because there’s no arbitrary execution.
I'm really intrigued by your point on read-memory vs a dedicated read interface, because it is a real insight about success rates in harness design.
How did you come to the conclusion you did? Could you speak a little to the evaluations you ran, or the data or anecdotes you collected to validate that decision?
I'm also curious about the overall framing of the question, which I'll challenge with, does the agent have to have a where?
An agent could be modeled by a set of states and transitions. I don't think that there's anything inherently necessary about the current "one process claude" approach for harnesses, other than convenience. Why hasn't a fully distributed harness, built on functions and tables, gained more mindshare?
-What remains unsolved is what should an Agent reasonably have access to in what context and for how long (etc).
Probabilistic code that can run far faster than human driven code, we don’t have a great model yet. We all should spend our energy there…
- Separating / putting controls on the FS resource is no different than putting the agent behind a firewall / allow-deny list.
It doesn’t invalidate running a sandbox in a sandbox to have better security.
I still kind of think it’s a decent idea but it’s too close to MCP with drawbacks that make it a harder sell than MCP. It’s hard to compete on functionality from a secure sandbox if users decide they don’t care about security.
On exe.dev the agent (Shelley) runs in a Linux VM, which is the security boundary. All the conversations are saved to a sqlite database, and it knows how to read it, so you can refer to a previous conversation in the database. It's also handy for asking the AI to do random sysadmin stuff, since it can use sudo.
A downside is that there's nowhere in the VM where secrets are safe from possibly getting exfiltrated via an injection attack. But they have "integrations" where you can put secrets into an http proxy server instead of having them locally.
Also, you don't need to use AI at all. You can use the VM as a VM.
But shouldn't there really be another sandbox where the agentic tool calls execute? This is to contain the damage of the tool execution when it goes wrong.
And, the agent harness itself should either implement or be contained in a third sandbox, which should contain the damage of the agent. There should be a firewall layer to limit what tool requests the agent can even make. This is to contain the damage of the agent when it formulates inappropriate requests.
The agent also should not possess credentials, so it cannot leak them to the LLM and allow them to be transformed into other content that might leak out via covert channels.
At the end of the day, it’s a “simple” loop that calls an external API (LLM) and receives requests to execute stuff on its behalf.
It’s not the agent running bash commands: you (the harness author) are, and you’re in full control of where and how those commands get executed.
In the article’s case, bash commands are forwarded to a sandbox, nothing ever runs on the harness itself (it physically can’t, local execution is not even implemented in the harness).
The reason agents work is because they have access to stuff by default. The whole world is context engineering at this point, and this proposal is to intermediate the context with a bespoke access layer. I put the bare minimum into getting my dev instance into a state where I can develop, because doing stuff (and these days: getting my agent to do stuff) is the goal.
This makes slightly more sense if you're building a SaaS and trying to get others to give you access to their code, their documents, and the rest so you can run agents against it. But the easiest, most powerful way is to just hook the agents up to the place that's already set up.
Tools, memories, sandboxing, steering, etc
Because there isn't really much more to it. And ever since we, i.e. those of us who played with ChatGPT API early on, bolted tools to it, some half a year before OpenAI woke up and officially named it "function calling" - ever since then, we knew that harness was the key. What kept changing was which logic (and how much of it) to put in explicitly, vs. pushing it back to the model on the "main thread", vs. pushing it to a model on a separate conversation track. But the basic insight remains the same.
--
[0] - Well, today - until recently you'd call it a "runner" or "runtime".
AI companies would love if everything ran in their cloud, but arguably there are latency reasons or other reasons to run at least some stuff in your own computer
The harness is the part that makes the API calls, interacts with the user, makes the function calls, and keeps track of the conversation memory.
You can also use the LLM to summarize the conversation into a single shorter message so you get compaction. And instead of statically defining which functions are available to the LLM you can create an MCP server which allows the LLM to auto-discover functions it can call and what they do.
That’s the whole magic of something like Claude Code. The rest is details.
Personally, for me it embodies a level of autonomy. I define that as, an AI model with potential to interact with something external to itself based on its output, where that includes its own future behavior.
Arguably this is a feature not a bug. Conflict resolution forces the need for a process to come to agreement on a common source of truth - one of the reasons why most Git repos don’t allow users to push to main directly. Writing directly to a shared memory database seems like it would result in chaos and a host of side effects once the number of users scales.
1) It's still assuming agents have CLIs. This is a very developer-centric concept of agents, and doesn't map well to either consumer or enterprise agents that aren't primarily working with files. Skills, plans, TODO lists, and memory are good, but don't have to be modeled as raw file access. Many harnesses have tools for them.
2) It's talking about a singular sandbox. That's not good enough for prompt injection prevention, secure credential management, and limiting the blast radius of attacks.
Another benefit of moving the harness outside the sandbox is you get to avoid accidentally creating a massive distributed system and you therefore don't have to think so much about events/communication between your main API and your sandboxes.
- Easy single command CLI agent spawning with templates
- Automatic context transfer (i. e. a bit like git worktrees)
- Fully containerised, but remote (a bit like pods)
- Central, mitm-proxy zero trust authn/authz management (no keys or credentials inside the agents), rather enrichment in the hypervisor/encapsulation
- Multi agent follow-up functionalities
- Fully self hosted/FOSS
Basically a very dev-friendly, secure, "kubernetes"-like solution for running remote agents.
Anyone has an idea of how to achieve this or potential technologies?
https://github.com/imbue-ai/mngr
This problem is quite common and not limited to memories. For instance, Claude Code will block write attempts and steer the agent to perform a read first (because the file might have been modified in the meantime by the user or another agent).
Same principle here: rather than trying to deterministically “merge” concurrent writes, you fail the last write and let the agent read again and try another write
Anyway. General advice: treat harnesses as any other (third-party) software that you run on your server. Modern harnesses (the ones from big companies, you need to subscribe to) are black boxes. Would you run a random binary you fetched from the internet on your server? Claude code, codex etc. are exactly this.