---
id: pydantic-monty-runtime
version: "0.0.21"
license: MIT
license_treatment: permissive
maintenance: active
---

# pydantic-monty-runtime — The monty CLI binary — spawned as worker subprocesses by pydantic-monty

License: permissive · Maintenance: active · Downloads: 328.0K/mo

## What it is and what it does

pydantic-monty-runtime is a compiled binary package that installs the `monty` command-line tool—a sandboxed Python interpreter written in Rust. It can run Python code in three modes: interactive REPL, file execution, or as a subprocess worker driven by a parent process. The sandbox enforces resource limits (memory, execution duration, recursion depth, garbage collection intervals) and provides crash isolation via the monty-alloc allocator, which raises MemoryError on soft-limit crossing and exits with a dedicated status on hard-limit crossing.

The package is primarily designed as a runtime dependency for pydantic-monty and pydantic-monty-client, which use it to execute untrusted or resource-constrained Python code safely. It supports optional type checking via ty, filesystem mounts with configurable access modes, and optional telemetry via Logfire (behind a feature flag). The binary is packaged for PyPI using the same model as uv and ruff, placing the compiled executable in the environment's scripts directory.

Use it for:

- Run untrusted Python code in a controlled sandbox with enforced memory and time limits
- Execute Python scripts as isolated worker subprocesses with crash recovery
- Type-check Python code before execution using the -t flag
- Mount host directories into the sandbox for controlled file access with ro/rw/overlay modes
- Integrate sandboxed Python execution into larger systems via pydantic-monty or monty-pool

## Worth the install?

AI-flagged interpretation of the facts above — verify before relying.

Provides the `monty` command-line binary, a sandboxed Python interpreter that runs Python code with resource limits and crash isolation, typically spawned as worker subprocesses by pydantic-monty.

Yes, if you need sandboxed Python execution with resource limits and crash isolation. Install directly only if you are building on top of pydantic-monty or monty-pool; otherwise, install pydantic-monty (which pulls it in automatically). No known vulnerabilities, active maintenance, and permissive MIT license. Medium install friction is offset by precompiled wheels across platforms.

## Install

pip install pydantic-monty-runtime
uv add pydantic-monty-runtime
poetry add pydantic-monty-runtime

## Installing pydantic-monty-runtime

Before you install: Medium install friction due to precompiled wheels across multiple platforms and Python versions (3.10+). Actively maintained with recent releases; intended as a dependency pulled in by pydantic-monty rather than direct installation.

License in practice: MIT license permits permissive use, modification, and distribution with minimal restrictions, making it suitable for both open-source and commercial projects.

Quickstart:

pip install pydantic-monty-runtime

# Then use the monty binary from the command line:
# monty -c "print('hello world')"
# monty file.py
# monty --max-memory 10MB --max-duration 0.5 -c "code"

Requires Python >= 3.10. The binary is precompiled; installation places it in the environment's scripts directory and assumes a compatible platform (macOS, Linux, or Windows).

Verify before relying:

- Whether the package works correctly when installed directly vs. only as a transitive dependency of pydantic-monty
- Performance characteristics and overhead of the sandbox relative to standard Python execution
- Completeness of stdlib subset implemented by the Monty interpreter

## Package facts

- License: MIT (permissive)
- Python support: supports_current
- Install friction: medium
- Maintenance: active
- Downloads: 328.0K/month (top 15,000 on PyPI)
- Known vulnerabilities: none known

## Tags

sandboxed python interpreter, python sandbox runtime, monty cli binary, resource-limited python execution, crash-isolated python worker, python code sandbox, restricted python environment, sandbox, resource-limits, crash-isolation

[View on SkillFed](https://skillfed.io/packages/pydantic-monty-runtime) · [View on PyPI](https://pypi.org/project/pydantic-monty-runtime/)