AI Agent Analysis

humanlayer/12-factor-agents

What are the principles we can use to build LLM-powered software that is actually good enough to put in the hands of production customers?

25k starsTypeScriptGitHub

Source and compliance noteLast synced: Jul 21, 2026

Git-Stars is independent and not affiliated with GitHub or this project. Analysis may be AI-assisted and based on public repository metadata plus short README-derived summaries. We do not mirror full README files, docs, issues, or social comments.

Original GitHub source Methodology Editorial Policy

Overview

12-Factor Agents is a set of principles for building reliable, production-grade LLM-powered software. It emphasizes deterministic code with LLM steps sprinkled in at the right points, rather than fully autonomous agent loops, drawing inspiration from the 12 Factor App methodology.

Installation

No installation required; the project is a guide. For the CLI tool, run `npx create-12-factor-agent` or `uvx create-12-factor-agent`.

What You Can Build

Developers can build customer-facing AI applications that are reliable, scalable, and maintainable, such as customer support bots, code assistants, or data processing pipelines. The ceiling is high: any application where LLM capabilities are needed but must be tightly controlled and integrated with existing software engineering practices.

Community Sentiment

Positive