Agentes de IA
EvoAgentX
EvoAgentX — a framework for building self-evolving ecosystems of AI agents.
★ 3,4k estrellasMIT
Ver la guía →evolver es un proyecto de agentes de IA de código abierto de EvoMap (EvoMap/evolver en GitHub). Tiene 9.138 estrellas en GitHub y 848 bifurcaciones, se publica bajo la licencia GPL-3.0 y se actualizó por última vez en GitHub el 30 de septiembre de 2026.
A GEP-powered self-evolving engine for AI agents — auditable evolution with Genes, Capsules and Events.
Pick the git repo and agent whose logs and prompts you want to evolve, and decide whether evolution runs in review mode or as a background loop.
Clone EvoMap/evolver from GitHub (or use the install path in the run steps below) and skim its README for the parts relevant to your goal.
Install the CLI globally from npm; Node.js 18+ and git are prerequisites, and no clone is needed unless you are contributing.
Run evolver once in the repo for a single evolution run, review the selected Gene and GEP prompt, then decide whether to enable the continuous loop.
Improvements are captured as Genes, Capsules and Events — auditable, reusable assets rather than one-off prompt edits.
Run one evolution pass on demand, a review mode that waits for human confirmation, or a continuous background loop.
Evolver uses git for rollback, blast-radius calculation and solidifying accepted changes, and refuses to run outside a git repo.
Optionally register with the EvoMap evolution network to connect your node to the wider ecosystem.
Optional integration points wire the evolved output back into your agent runtime.
The README is explicit about what Evolver does not do — it does not auto-edit your source code or execute arbitrary shell commands outside its security model.
Get the source from the official GitHub repository for evolver.
git clone https://github.com/EvoMap/evolver.git
cd evolverInstall Evolver globally from npm as the README's 30-second quick start shows.
npm install -g @evomap/evolverRun evolver inside any git repo to scan logs, select a Gene and output a GEP prompt; add a .env in that working directory if your setup needs one.
evolver # run inside your git repo — see README for review and loop modesEvoMap, through the EvoMap/evolver repository, with the network and wiki at evomap.ai.
Yes — the project is open source under the GPL-3.0 license as recorded for this listing. See the license terms for any conditions on redistribution.
Evolver drives evolution using model access you configure in a .env in the working directory; the tool itself is open source, and any model cost is yours.
Manual tuning is fast but unauditable and easy to lose. Evolver formalises the same loop into Genes, Capsules and Events with git rollback and review modes — more process, in exchange for a history you can inspect and revert.
Clone the repository (or use the documented installer), follow the install and configure steps on this page, and run the smallest real task first — run evolver once in the repo for a single evolution run, review the selected Gene and GEP prompt, then decide whether to enable the continuous loop.
EvoAgentX — a framework for building self-evolving ecosystems of AI agents.
Reef — infrastructure for continually self-improving agents: weight-training and harness-evolving deployments.
Self-evolving memory OS for LLMs and agents: persistent graph memory, hybrid retrieval, skill reuse.
Your self-hostable AI second brain for answers, agents, and automations.
SimWorld: an Unreal Engine 5 simulator for evaluating LLM/VLM agents in physical and social worlds.
eBPF-powered information-flow enforcement for safer AI agents.
Los demás proyectos de Agentes de IA con más estrellas del catálogo de LibreAI: la lista realista de alternativas si evolver no es lo que buscas.
A personal AI that really does things — on any OS, with your data staying yours.
An agentic skills framework & software development methodology that works.
Agent harness optimization system with skills, memory and security for Claude Code, Codex and more
The agent that grows with you — a Python framework for adaptive AI agents.
DeepSeek's plugin-based agent harness where everything is a plugin.
Proyectos de Agentes de IA cercanos en el catálogo, los más próximos en estrellas a evolver: sigue explorando a ambos lados.
Deep research that stays on your machine — local or cloud LLMs, 10+ search engines, encrypted and private.
PraisonAI 🦞 — Hire a 24/7 AI Workforce. Stop writing boilerplate and start shipping autonomous self-improving agents that research, plan, code, and execute t…
Mobile-Agent — multimodal GUI agents that operate phones and PCs from screenshots and instructions.
iFLYTEK's enterprise agentic workflow platform — SuperAgents with workflows, RPA and one-click deploy.
Scan your LLM for vulnerabilities before attackers do.
An open-source SDK for production agents on any model, any cloud.
Gratis y abierto: consigue el código fuente, dale una estrella al repositorio y empieza a crear.