Features
Provision Postgres, auth, object storage, realtime, edge functions per agent
MicroVM sandboxes for agent runtimes (Hermes, OpenClaw, Codex, Claude Code)
Declarative agent config file (naive.config.ts) with versioning
/compute: run Docker containers, workers, cron jobs
/queue: durable SQS-compatible queues (standard, FIFO, DLQ)
/phone: provision US phone numbers for SMS with 10DLC
/trading: trade stocks, options, crypto via OAuth brokerage
Model routing across 300+ models with smart task routing and automatic fallbacks
Governance: audit log, spend caps, approval gates, absolute revoke
Scoped virtual Visa cards with hard spend limits and invoicing
Identity: KYC, KYB, US LLC/EIN, email, phone per agent
Multi-agent orchestration runtime from single config
CLI, SDK (Node), and MCP support for agent control
1 API key works across SDK, CLI, and MCP
Templates for AI SEO, voice receptionist, autonomous company, mobile-use agents
Decorator-based LLM call pattern (@llm.call)
Automatic versioning with @ops.version
Tracing and cost tracking per call via @ops.trace
Built-in tool calling with @llm.tool
Agent loop with tool execution and response resume
Streaming response support
Thinking/reasoning with include_thoughts parameter
OpenTelemetry integration for observability
Structured output via Python type hints
Multi-provider support (OpenAI, Anthropic, Google)
Lightweight, minimal abstraction design
Provider-agnostic unified interface
Simple tool creation with function decorators
Cost tracking per interaction
GitHub open-source (MIT license)
