π The original agent harness. Deploy intelligent multi-player swarms, coordinate autonomous workflows, and build conversational AI systems. Features adaptive memory, self-learning intelligence, federation, vector RAG integration, and native Claude Code / Codex / Hermes and many more Integrated
TypeScript agentic-aiagentic-frameworkagentic-workflowagentsai-agentsai-assistantai-skillsautonomous-agentsclaudecodexdsh-pluginharnessmcp-servermulti-agentmulti-agent-systemsnpmskillsswarmswarm-intelligencetypescript
Checkpoint: File edits
rUvbc12e76dbf1a
diff --git a/README.md b/README.mdindex 8a4cbb38c5..3f95da03e0 100644--- a/README.md+++ b/README.md@@ -179,22 +179,22 @@ Claude-Flow v3 introduces **self-learning neural capabilities** that no other ag| Feature | Claude Flow v3 | CrewAI | LangGraph | AutoGen | MetaGPT ||---------|----------------|--------|-----------|---------|---------|-| **Self-Learning** | SONA + EWC++ | π΄ | π΄ | π΄ | π΄ |-| **Prevents Forgetting** | EWC++ consolidation | π΄ | π΄ | π΄ | π΄ |-| **Expert Routing** | MoE (8 experts) | Manual roles | Graph edges | π΄ | Fixed roles |-| **Vector Memory** | HNSW (150x faster) | π΄ | Via plugins | π΄ | π΄ |-| **Hyperbolic Embeddings** | PoincarΓ© ball model | π΄ | π΄ | π΄ | π΄ |-| **Pattern Learning** | From trajectories | π΄ | π΄ | π΄ | π΄ |-| **Work Ownership** | Claims system | π΄ | π΄ | π΄ | π΄ |-| **Threat Detection** | AIDefence | π΄ | π΄ | π΄ | π΄ |-| **Attention Optimization** | Flash Attention | π΄ | π΄ | π΄ | π΄ |-| **Low-Rank Adaptation** | LoRA (128x compression) | π΄ | π΄ | π΄ | π΄ |-| **Quantization** | Int8 (3.92x savings) | π΄ | π΄ | π΄ | π΄ |-| **Consensus Protocols** | 5 (Raft, Byzantine, etc.) | π΄ | π΄ | π΄ | π΄ |-| **Background Workers** | 12 auto-triggered | π΄ | π΄ | π΄ | π΄ |-| **Multi-Provider LLM** | 6 with failover | 2 | 3 | 2 | 1 |-| **MCP Integration** | Native | π΄ | π΄ | π΄ | π΄ |-| **Swarm Topologies** | 4 (mesh, hierarchical, etc.) | 1 | 1 | 1 | 1 |+| **Self-Learning** | β SONA + EWC++ | π΄ | π΄ | π΄ | π΄ |+| **Prevents Forgetting** | β EWC++ consolidation | π΄ | π΄ | π΄ | π΄ |+| **Expert Routing** | β MoE (8 experts) | Manual roles | Graph edges | π΄ | Fixed roles |+| **Vector Memory** | β HNSW (150x faster) | π΄ | Via plugins | π΄ | π΄ |+| **Hyperbolic Embeddings** | β PoincarΓ© ball model | π΄ | π΄ | π΄ | π΄ |+| **Pattern Learning** | β From trajectories | π΄ | π΄ | π΄ | π΄ |+| **Work Ownership** | β Claims system | π΄ | π΄ | π΄ | π΄ |+| **Threat Detection** | β AIDefence | π΄ | π΄ | π΄ | π΄ |+| **Attention Optimization** | β Flash Attention | π΄ | π΄ | π΄ | π΄ |+| **Low-Rank Adaptation** | β LoRA (128x compression) | π΄ | π΄ | π΄ | π΄ |+| **Quantization** | β Int8 (3.92x savings) | π΄ | π΄ | π΄ | π΄ |+| **Consensus Protocols** | β 5 (Raft, Byzantine, etc.) | π΄ | π΄ | π΄ | π΄ |+| **Background Workers** | β 12 auto-triggered | π΄ | π΄ | π΄ | π΄ |+| **Multi-Provider LLM** | β 6 with failover | 2 | 3 | 2 | 1 |+| **MCP Integration** | β Native | π΄ | π΄ | π΄ | π΄ |+| **Swarm Topologies** | β 4 (mesh, hierarchical, etc.) | 1 | 1 | 1 | 1 |**Key differentiators:**- **SONA (Self-Optimizing Neural Architecture)** - Learns optimal agent routing from trajectory outcomes with <0.05ms adaptation