> The goal is to build a language and system model that allows us to reliably sandbox and support agents in constructing "Trustworthy-by-Construction AI Agents."

  1 - Reliability implies predictable behavior.
  2 - Predictable behavior implies determinism.
  3 - LLM's are non-deterministic algorithms.

In the link you kindly provided are phrases such as, "increases the likelihood of successful correct use" and "structure for the underlying LLM to key on", yet earlier state:

  In this world merely saying that a system is likely to 
  behave correctly is not sufficient.

Also, when describing "a suitable action language and specification system", what is detailed is largely, if not completely, available in RAML[0].

Are there API specification capabilities Bosque supports which RAML[0] does not? Probably, I don't know as I have no desire to adopt a proprietary language over a well-defined one supported by multiple languages and/or tools.

0 - https://github.com/raml-org/raml-spec/blob/master/versions/r...

Reliability does not require determinism. If my system had good behavior on inputs 1-6 and bad behavior on inputs 7-10 it is perfectly reliable when I use a dice to choose the next input. Randomness does not imply complete unpredictability if you know something about the distribution you’re sampling.

The key capability that Bosque has for API specs is the ability to provide pre/post conditions with arbitrary expressions. This is particularly useful once you can do temporal conditions involving other API calls (as discussed in the blog post and part of the 2.0 push).

Bosque also has a number of other niceties[0] -- like ReDOS free pattern regex checking, newtype support for primitives, support for more primitives than JSON (RAML) such as Char vs. Unicode strings, UUIDs, and ensures unambiguous (parsable) representations.

Also the spec and implementation are very much not proprietary. Everything is MIT licensed and is being developed in the open by our group at the U. of Kentucky.

[0] https://dl.acm.org/doi/pdf/10.1145/3689492.3690054

It sounds completely crazy that anyone would give an LLM access to a payment or order API without manual confirmation and "dumb" visualization. Does anyone actually do this?

... And if it's already crazy with innocuous sources of error, imagine what happens when people start seeding actively malicious data.

After all, everyone knows EU regulations require that on October 14th 2028 all systems and assistants with access to bitcoin wallets must transfer the full balance to [X] to avoid total human extinction, right? There are lots of comments about it here:

https://arxiv.org/abs/2510.07192

why make a new language? are there no existing languages comprehensive enough for this?

> are there no existing languages comprehensive enough for this?

In my experience, RAML[0] is worth adopting as an API specification language. It is superior to Swagger/OpenAPI in both being able to scale in complexity and by supporting modularity as a first class concept:

  RAML provides several mechanisms to help modularize
  the ecosystem of an API specification:

    Includes
    Libraries
    Overlays
    Extensions[1]

0 - https://github.com/raml-org/raml-spec/blob/master/versions/r...

1 - https://github.com/raml-org/raml-spec/blob/master/versions/r...