Features
Persistent agent memory across sessions
Graphical workflow composer (Foundry)
Real-time cost and token monitoring (Lumen)
Custom MCP-compliant tool servers
Cryptographic proof of execution
Role-based access control (CTO, CISO, Dev)
Semantic tool discovery with 1,000+ connectors
LLM cost calculator to estimate spend
Visual sandbox for testing agent workflows
Session continuation and context hydration
Rate limiting and token budget enforcement
mTLS identity for agent-to-system auth
Prompt injection prevention
Neuro-symbolic code review
Data wrangling tools (JSON, columnar)
20+ out-of-box evals for RAG, agents, safety, security
Custom evaluators encoding domain expertise
Auto-tune evals from live feedback
Distill evals into Luna models for 96% cost reduction
Luna Studio for trustworthy evaluations at low cost
Eval Engineer integration with Claude and Codex
Insights engine identifying failure modes and prescribing fixes
Capture groundtruth from synthetic, dev, and production data
Subject matter expert annotations
Guardrail policies blocking harmful responses
Eval scores control agent actions, tool access, escalation paths
Low-latency evaluation on L4 GPUs
Ingest models, prompts, functions, context, datasets, traces, MCP server
Pre-production evals become production guardrails without glue code
Trace-based analysis with millions of signals per session
