Show HN: The Mog Programming Language (moglang.org)
- Mog is a statically typed, compiled, embedded language (think statically typed Lua) designed to be written by LLMs -- the full spec fits in 3,200 tokens. - An AI agent writes a Mog program, compiles it, and dynamically loads it as a plugin, script, or hook. - The host controls exactly which functions a Mog program can call (capability-based permissions), so permissions propagate from agent to agent-written code. - Compiled to native code for low-latency plugin execution -- no interpreter overhead, no JIT, no process startup cost. - The compiler is written in safe Rust so the entire toolchain can be audited for security. Even without a full security audit, Mog is already useful for agents extending themselves with their own code. - MIT licensed, contributions welcome.
Motivations for Mog:
1. Syntax Only an AI Could Love: Mog is written for AIs to write, so the spec fits easily in context (~3200 tokens), and it's intended to minimize foot-guns to lower the error rate when generating Mog code. This is why Mog has no operator precedence: non-associative operations have to use parentheses, e.g. (a + b) * c. It's also why there's no implicit type coercion, which I've found over the decades to be an annoying source of runtime bugs. There's also less support in Mog for generics, and there's absolutely no support for metaprogramming, macros, or syntactic abstraction.
When asking people to write code in a language, these restrictions could be onerous. But LLMs don't care, and the less expressivity you trust them with, the better.
2. Capabilities-Based Permissionsl: There's a paradox with existing security models for AI agents. If you give an agent like OpenClaw unfettered access to your data, that's insecure and you'll get pwned. But if you sandbox it, it can't do most of what you want. Worse, if you run scripts the agent wrote, those scripts don't inherit the permissions that constrain the agent's own bash tool calls, which leads to pwnage and other chaos. And that's not even assuming you run one of the many OpenClaw plugins with malware.
Mog tries to solve this by taking inspiration from embedded languages. It compiles all the way to machine code, ahead of time, but the compiler doesn't output any dangerous code (at least it shouldn't -- Mog is quite new, so that could still be buggy). This allows a host program, such as an AI agent, to generate Mog source code, compile it, and load it into itself using dlopen(), while maintaining security guarantees.
The main trick is that a Mog program on its own can't do much. It has no direct access to syscalls, libc, or memory. It can basically call functions, do heap allocations (but only within the arena the host gives it), and return something. If the host wants the Mog program to be able to do I/O, it has to supply the functions that the Mog program will call. A core invariant is that a Mog program should never be able to crash the host program, corrupt its state, or consume more resources than the host allows.
This allows the host to inspect the arguments to any potentially dangerous operation that the Mog program attempts, since it's code that runs in the host. For example, a host agent could give a Mog program a function to run a bash command, then enforce its own session-level permissions on that command, even though the command was dynamically generated by a plugin that was written without prior knowledge of those permission settings.
(There are a couple other tricks that PL people might find interesting. One is that the host can limit the execution time of the guest program. It does this using cooperative interrupt polling, i.e. the compiler inserts runtime checks that check if the host has asked the guest to stop. This causes a roughly 10% drop in performance on extremely tight loops, which are the worst case. It could almost certainly be optimized.)
3. Self Modification Without Restart: When I try to modify my OpenClaw from my ph...
40 comments
[ 2.7 ms ] story [ 62.1 ms ] threadI see that Deno requires a subprocess which introduces some overhead, and I might be naive to think so, but that doesn't seem like it would matter much when agent round-trip and inference time is way, way longer than any inefficiency a subprocess would introduce. (edit: I realized in some cases the round-trip time may be negligible if the agent is local, but inference is still very slow)
I admittedly do prefer the syntax here, but I'm more so asking these questions from a point of pragmatism over idealism. I already use Deno because it's convenient, practical, and efficient rather than ideal.
[1] https://www.merriam-webster.com/slang/mog
If you're running the compiled code in-process, how is that not JIT? And isn't that higher-latency than interpreting? Tiered-JIT (a la V8) solves exactly this problem.
Edit: Although the example programs show traditional AOT compile/execute steps, so "no process startup cost" is presumably a lie?
Please think twice before releasing these, if you're going to do it come up with at least one original idea that nobody else has done before.
Why didn't you just call it "bad rust copy"?
A few questions:
- Is there a list of host languages?
- Can it live in the browser? (= is JS one of the host languages?)
But LLMs very much do care. They are measurably worse when writing code in languages with non-standard or non-existent operator precedence. This is not surprising given how they learn programmming.
Almost all the code LLMs have been trained on uses operator precedence, so no operator precedence seems like a massive foot-gun.
> There's also less support in Mog for generics, and there's absolutely no support for metaprogramming, macros, or syntactic abstraction.
OK that does immediately make it boring, I give them that much.
Agents can pretty much iterate on their own.
The most important thing for me, at least for now (and IMO the foreseeable future) is being able to review and read the output code clearly. I am the bottleneck in the agent -> human loop, so optimizing for that by producing clear and readable code is a massive priority. Gleam eliminates a ton of errors automatically so my reviews are focused on mostly business logic (also need to explicitly call out redundant code often enough).
I could see an argument for full on Erlang too, but I like the static typing.
Would have been a blockchain language 10 years ago.
> String Slicing > You can extract a substring using bracket syntax with a range: s[start:end]. Both start and end are byte offsets. The slice includes start and excludes end.
Given that all strings are UTF-8, I note that there's not a great way to iterate over strings by _code point_. Using byte offsets is certainly more performant, but I could see this being a common request if you're expecting a lot of string manipulation to happen in these programs.
Other than that, this looks pretty cool. Unlike other commenters, I kinda like the lack of operator precedence. I wouldn't be surprised if it turns out to be not a huge problem, since LLMs generating code with this language would be pattern-matching on existing code, which will always have explicit parentheses.