# lealone/Lealone

能安全适用于氛围编程和企业应用的全栈自进化通用智能体

Repository: https://github.com/lealone/Lealone
Canonical: https://ross.abutalabs.com/products/lealone
Language: Java
License: NOASSERTION
License Family: other
Topics: newsql, rdbms, database, async, sharding, replication, acid, orm, oltp, lealone, microservice, sql, agent, ai, llm
Last push: 2026-08-26T00:23:00+00:00

## Health v2 (maintenance only)
Score: 67/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 99, release rhythm 8, longevity 100
- inputs: {"age_days": 4985, "days_push": 8, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: no_license
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 2581, forks 522 (observed 2026-08-28T04:07:02.425924+00:00)

## What it is
Lealone is a Java-based full-stack, self-evolving general-purpose AI agent built on top of a high-performance NewSQL/RDBMS engine with async, sharding, replication, and ACID support. It lets users describe applications or services in natural language or SQL-like syntax (create service, create workflow) and generates runnable enterprise applications with HTTP endpoints.

## Use cases
- build a todo app from a natural language prompt (vibe coding)
- create and execute backend services with SQL-like syntax
- run an enterprise AI application from a requirements document
- prototype full-stack apps without writing code or requirement docs
- use a distributed NewSQL database with sharding and replication
- build microservices with built-in ORM and async database access

## When to choose
- you want to generate full-stack applications from natural language prompts
- you need an embeddable Java NewSQL database with async, sharding, and ACID guarantees
- you want to define services and workflows declaratively in SQL-like syntax
- you are prototyping enterprise apps quickly with LLM assistance

## When to avoid
- you need a battle-tested production database with a large community and proven track record
- you require support for LLM providers other than doubao or deepseek
- you need a non-JVM stack or fine-grained control over generated code
- you prefer traditional explicit coding over AI-driven generation

## Facets
- artifact type: framework
- maturity: active
- function: database, orm, agent-framework, llm-inference, web-framework, rpc, api-framework
- domain: databases, artificial-intelligence, large-language-models, web-development, backend, developer-tools
- platform: jvm, cross-platform, cli
- tags: newsql, rdbms, vibe-coding, self-evolving-agent, sharding, replication, acid, oltp, microservices, sql-services, async, ai-agents, web-server

## Member repositories
- lealone/Lealone (main) score 67

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:07:02.425924+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 2026-08-30T02:22:37.207720+00:00, confidence not recorded.
  - readme: https://github.com/lealone/Lealone (fetched 2026-08-28T04:07:02.425924+00:00, sha 5d7a778e6afa)
- Data as of 2026-08-30T08:39:29.467469+00:00.