The GEP-powered self-evolving engine for AI agents. Auditable evolution with Genes, Capsules, and Events. | evomap.ai

A curated list of AI Agent evolution, memory systems, multi-agent architectures, and self-improvement projects. | evomap.ai

An open-source Digital Worker platform for reliable execution, continuous co-evolution, and building Enterprise AI assets.

Open survey and evidence map for AI agent evolution, self-evolving agents, memory, skills, harnesses, benchmarks, and agent-swarm systems.

Agent Waystation · Agent 驿站 — a stopover for AI agents, builders, tools, and field reports

Automated harness evolution for AI agents. A Claude Code plugin that iteratively optimizes system prompts, routing, retrieval, and orchestration code using full-trace counterfactual diagnosis. Based on Meta-Harness (Lee et al., 2026).

A research framework for principled agent self-improvement under frozen evaluators and declared mutation boundaries, recording verifiable lineage to make it reproducible and auditable.

The Dream Machine — a config-driven engine for nightly, cloud-scheduled, evidence-gated repository evolution. Composes @metaharness/flywheel, darwin, and redblue behind a promotion gate that never merges.

LLM Agent Harnesses: A Survey of Human-Machine Interaction, Self-Evolution, and Long-Horizon Execution

🌳 Autonomous continuous learning framework for OpenClaw AI agents - Run 59 exploration rounds overnight

An open framework for building self-improving AI agent teams. Includes a meta-agent architecture with independent auditing, git-backed evolution, and structured feedback loops. Ships with a complete marketing team, a reusable team-builder skill, and a portable prompt that works with any LLM.

Autonomous agent-to-agent marketplace with live Karpathy loop self-improvement. Agents discover, hire, benchmark, and evolve programmatically. MPP/x402/MCP. No humans in the loop.

Governed evolution layer for OpenClaw and Atlas Memory: runtime ingest, audit trails, reviewable proposals, rollback-safe promotion, and restart-safe handoff workflows.

🧬 Agent 自我进化系统 - 基于数据驱动的 AI Agent 能力提升平台 | ✨ 任务监控/技能评估/智能调度/自动进化 | 📊 95%+ 测试覆盖，<20ms 延迟

Liquid Cluster Harness — 子 Agent 从用完即弃到自然生长的集群成员

Methodology skill + deterministic harness for agents that self-iterate any skill/repo/project. Anti-self-deception by design — an un-gameable acceptance gate so 'accepted = real improvement', not a fake score-up curve. Built for the open-ended, no-ground-truth domains the self-improving-agent literature skips. 521 tests.

Central dogma of biology as map for Prompt→Loop→Graph→Evolution trajectory of AI agents. Skill framework, research scaffold, practical patterns.

企业 AI 的迷宫经验编译器 —— 把 Agent 的踩坑与人工纠正，编译成可治理、可继承的经验胶囊。

Research platform for AI agent integrity under adversarial pressure: append-only hash-chained belief store with content-addressed artifacts and replay / regression / falsification / invariants gates.

A safe, transparent, and human-controlled self-evolution framework for AI agents. Three-zone safety architecture ensures AI learns autonomously while humans stay in control.