Skip to main content

Python that lacks most of Python. Restricted program generation and execution for tool composition.

Project description

lackpy

Python that lacks most of Python. A micro-inferencer that translates natural language intent into sandboxed tool-composition programs.

PyPI Tests Docs

What it does

$ lackpy -c "read file main.py and count its lines" --kit read_file,find_files

lackpy takes an intent, generates a restricted Python program using a local model you configure (any Ollama-served coder model), validates it against a strict AST whitelist, and runs it with traced tool calls. One MCP call replaces N tool round-trips.

Install

pip install lackpy            # runtime + local/cloud model inference (via woollama)
pip install lackpy[full]      # + all optional features (sandbox, mcp, blq, …)

pip install lackpy-lang       # JUST the language (grammar/validator/grader/spec),
                              # no runtime — for tools that only reason about programs

Model inference is built in — calls route through woollama's core (a dependency), so local Ollama and cloud backends (Anthropic, OpenAI, …) all work without a per-vendor extra; cloud backends just need the relevant API key in your environment.

lackpy and lackpy-lang share the lackpy namespace (PEP 420); neither ships a top-level lackpy/__init__.py, so import from submodules (from lackpy.lang import validate, from lackpy.service import LackpyService).

Development (monorepo)

This repo builds two distributions. For a working editable checkout, install bothlackpy pins lackpy-lang, which isn't on PyPI until its own tag, so install the local copy first:

pip install -e ./packages/lackpy-lang -e ".[dev]"

Quick start

lackpyctl init --ollama-url http://localhost:11434   # writes .lackpy/config.toml
lackpy -c "find all python files" --kit read_file,find_files

lackpyctl init wires a local Ollama model into the inference order — pick the model with --ollama-model (the choice is per-machine, not a package default). Without it, only the deterministic templates/rules tiers run and compositional intents won't generate.

from lackpy.service import LackpyService
from lackpy.kit.toolbox import ToolSpec, ArgSpec
import asyncio

async def main():
    svc = LackpyService()
    result = await svc.delegate("read file README.md", kit=["read_file"])
    print(result["output"])

asyncio.run(main())

Documentation

Full docs at lackpy.readthedocs.io:

Part of the Rigged Suite

lackpy is part of the Rigged developer tool suite but is independently installable.

License

MIT

Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

lackpy-0.12.0.tar.gz (533.3 kB view details)

Uploaded Source

Built Distribution

If you're not sure about the file name format, learn more about wheel file names.

lackpy-0.12.0-py3-none-any.whl (159.6 kB view details)

Uploaded Python 3

File details

Details for the file lackpy-0.12.0.tar.gz.

File metadata

  • Download URL: lackpy-0.12.0.tar.gz
  • Upload date:
  • Size: 533.3 kB
  • Tags: Source
  • Uploaded using Trusted Publishing? Yes
  • Uploaded via: twine/6.1.0 CPython/3.13.12

File hashes

Hashes for lackpy-0.12.0.tar.gz
Algorithm Hash digest
SHA256 5e09e8e025a02583cedce1425eae6b44d908a3b5967393d9bc65f0ba1eda769b
MD5 3a430deb74967df0681afc2ed127837e
BLAKE2b-256 fa5dfaedaa38d14e2da008c45ced70da46a69837a17d041cd506ab2bcaca0364

See more details on using hashes here.

Provenance

The following attestation bundles were made for lackpy-0.12.0.tar.gz:

Publisher: ci.yml on teaguesterling/lackpy

Attestations: Values shown here reflect the state when the release was signed and may no longer be current.

File details

Details for the file lackpy-0.12.0-py3-none-any.whl.

File metadata

  • Download URL: lackpy-0.12.0-py3-none-any.whl
  • Upload date:
  • Size: 159.6 kB
  • Tags: Python 3
  • Uploaded using Trusted Publishing? Yes
  • Uploaded via: twine/6.1.0 CPython/3.13.12

File hashes

Hashes for lackpy-0.12.0-py3-none-any.whl
Algorithm Hash digest
SHA256 9a5716295ea7d2b5987fc91f44be421b1b18b1578e95fa64fe39feeb1acb809b
MD5 11a2c1b6a360f361361f2c825e83fa32
BLAKE2b-256 ee2e2095c1958dd32a1d76947933cdebca74019d7000789a02f0ee9d1960e159

See more details on using hashes here.

Provenance

The following attestation bundles were made for lackpy-0.12.0-py3-none-any.whl:

Publisher: ci.yml on teaguesterling/lackpy

Attestations: Values shown here reflect the state when the release was signed and may no longer be current.

Supported by

AWS Cloud computing and Security Sponsor Datadog Monitoring Depot Continuous Integration Fastly CDN Google Download Analytics Pingdom Monitoring Sentry Error logging StatusPage Status page