pydantic-monty-client

Python client for the Monty sandboxed Python interpreter.

Most users want pydantic-monty instead, which pulls in this package plus pydantic-monty-runtime and is documented in full on its PyPI page. Install this one directly only when the monty binary is supplied some other way — a base image, a system package, or a build of this repository:

pip install pydantic-monty-client

Locating the worker binary

Execution always happens in a pool of monty worker subprocesses: a monty process can never be made fully crash-proof against memory errors (stack overflows, allocator aborts) triggered by adversarial input, so crash isolation is built in. Without pydantic-monty-runtime installed, pydantic_monty has to find that binary itself, in this order:

1. the binary_path= argument to Monty(...) / AsyncMonty(...)
2. the MONTY_BIN environment variable
3. the environment's scripts directory (where pydantic-monty-runtime installs it)
4. a monty executable on PATH
5. a cargo-built binary in the monty workspace, for editable installs of this repo

If none match, constructing a pool raises FileNotFoundError. The binary need not be the same release as this package, but it must speak a compatible wire protocol version — the worker rejects an incompatible one when a session is checked out, reporting the range it serves.

Usage

from pydantic_monty import Monty

with Monty() as pool:
    with pool.checkout() as session:
        print(session.feed_run('1 + 2'))
        #> 3

See the pydantic-monty README for async usage, external functions, snapshots, resource limits, type checking and observability, and limitations/pool-architecture.md in the repository for the behavioural details of subprocess execution.