#104104th of 142 in Agent frameworks and SDKs
embabel-agent
JVM framework for agents that plan actions toward goals, in Kotlin or Java
Live demo ↗
(opens in a new tab)Documentation ↗
(opens in a new tab)Website ↗
(opens in a new tab)Repository on GitHub ↗
(opens in a new tab)
- Stars
- 4.6k
- License
- Apache-2.0
- Last commit
- Nov 2026
- Last release
- Oct 2026
- Language
- Kotlin
Missed the latest GitHub check; these facts may be older.
Overview
Embabel is a Kotlin framework, usable from Java, for writing agent flows on Spring Boot. Developers declare actions, goals and conditions over a typed domain model, and the platform plans the action sequence with Goal Oriented Action Planning (or Utility AI) and replans after each step. LLM calls and plain code mix in one flow, and each action can use a different model.
Who it is for: JVM and Spring developers building goal-driven LLM agents
Strengths
- Non-LLM planner (GOAP, Utility AI option) picks action order and replans after each action
- Typed domain model and Spring injection; unit and end-to-end agent testing supported
- Per-action model selection, so cheaper or local models can handle point tasks
- Java and Kotlin templates, an examples repo and the Tripper sample app
Weaknesses
- JVM and Spring only; no Python or TypeScript SDK
- Natural-language actions, evolving mode and federation are described as future work
- Open mode is the least deterministic and can choose paths developers did not foresee
- Quick start assumes an OpenAI API key and Maven; docs link is a 1.5.0-SNAPSHOT guide
What it needs
- no GPU
- Needs Maven, Spring Boot, OpenAI API key (quick start)
- Models: OpenAI, local models (unspecified)
Also in Agent frameworks and SDKs
See all 142| Rank | Project | Score |
|---|---|---|
| 1 | ppt-masterAgent skill that turns documents into natively editable PowerPoint decks | 96 out of 100 |
| 2 | openai-agents-pythonPython SDK for multi-agent workflows with handoffs, guardrails and tracing | 92 out of 100 |
| 3 | mastraTypeScript framework for building AI agents, workflows and MCP servers | 92 out of 100 |
| #4 | browser-usePython library that lets an LLM agent control a web browser | 90 out of 100 |
| #5 | copilotkitReact and Angular SDK for adding agent chat and generative UI to apps | 90 out of 100 |