Features
Multi-agent shared memory with fleet scoping
Built-in governance with visibility scopes, trust tiers, tenant isolation
Full audit trail on every operation
Knowledge graph with entity and relation extraction
Hybrid recall: semantic + keyword + graph expansion
Self-improving retrieval via per-agent tuning and outcome feedback
LLM enrichment pipeline (classify, entity extract, contradiction check, embed)
Skill Factory distills agent experiences into reusable governed skills
Keystone policies for deterministic enforcement of rules
MCP-native integration for any AI client
OpenClaw plugin for fleet deployments with OTA updates
12 MCP tools: write, recall, list, manage, doc, entity get, tune, insights, evolve, stats, keystones, keystones_set
PII guard with auto-detection and quarantine
Agent cold-start resolution via pre-seeded knowledge and mandatory keystones
SOC 2 compliant and Apache 2.0 open source
Distributed tracing for LLM agents
Capture prompts, retrievals, tool calls, outputs
Human annotations on traces
Create datasets from traces
Run experiments to compare changes
OpenTelemetry-native instrumentation
Self-host locally, Docker, or Kubernetes
Cloud instances with free tier
Vendor agnostic: any model or framework
3M+ monthly downloads, 9k+ GitHub stars
