# ctlllll/LLM-ToolMaker

Repository: https://github.com/ctlllll/LLM-ToolMaker
Canonical: https://ross.abutalabs.com/products/llm-toolmaker
Language: Jupyter Notebook
License Family: other
Last push: 2023-05-29T09:23:54+00:00

## Health v2 (maintenance only)
Score: 29/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 0, release rhythm 35, longevity 85
- inputs: {"age_days": 1196, "days_push": 1192, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: no_releases, no_license
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 1064, forks 97 (observed 2026-08-28T04:03:26.604149+00:00)

## What it is
Research code for the LATM (LLMs as Tool Makers) framework, where a powerful LLM writes reusable Python tool functions and a cheaper LLM uses them for problem-solving. It accompanies an arXiv paper and is validated on complex reasoning tasks including Big-Bench.

## Use cases
- reproduce the LATM tool-making experiment
- let an LLM create its own reusable tools
- reduce inference cost by pairing GPT-4 tool maker with GPT-3.5 tool user
- evaluate LLMs on Big-Bench reasoning tasks
- explore closed-loop tool generation for agents

## When to choose
- you want to experiment with or extend the LATM paper's approach
- you need a reference implementation of LLM tool creation and reuse
- you're researching cost-efficient LLM agent pipelines

## When to avoid
- you need a production-ready agent framework
- you want a maintained library with a license and support
- you need tool-making without access to OpenAI GPT models

## Facets
- artifact type: learning-resource
- maturity: experimental
- function: agent-framework, llm-inference, machine-learning
- domain: artificial-intelligence, large-language-models
- platform: python, cross-platform
- tags: latm, tool-making, llm-agents, research-paper, big-bench, reasoning, jupyter-notebook, code-generation, ai-agents, research

## Member repositories
- ctlllll/LLM-ToolMaker (main) score 29

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:03:26.604149+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-30T06:56:03.528591+00:00, confidence not recorded.
  - readme: https://github.com/ctlllll/LLM-ToolMaker (fetched 2026-08-28T04:03:26.604149+00:00, sha 523a6fd785be)
- Data as of 2026-08-30T08:39:29.467469+00:00.