Spend controls, budgets and approvals for AI agents
Production control plane for AI agents with model gateway, stateful runtime, and observability
Stage
MVP
Not given
Founded
Not given
Not given
Based in
Not given
Not given
Pricing model
Freemium
Paid
Pricing
Free plan: $0 during early access with 2 seats and 3 agents. Planned paid plans: Builder at $49/month with 5 seats and 25 agents; Scale at $199/month with 25 seats and 250 agents. Self-hosting is free.
Relay: usage by model through API. Loop: base + usage for agent runtime capacity and managed runs. Lens: base + usage for observability retention and recorded events. Build Agent Service: custom scoped professional service. WebSearch API: $0.01 per successful simple/reader request, $0.08 per success
Key features
Onchain allowances and epoch budgets
Escalation through reporting tree
Session keys with automatic expiration
Constitutional governance with veto windows
Declared flows and human gates
Slack and Telegram supervision
Visual org builder and workspace
Apache-2.0 open source contracts and SDK
Inference budgets for model and API calls
Integration with Coinbase AgentKit and Safe
Model protocol translation for OpenAI, Anthropic, and Gemini SDKs
Policy routing and approval gates for model calls
Stateful agent runtime with session management
Managed tool attachment and control
Decision replay and audit trails
WebSearch API with multiple query types
Full request traceability from model to outcome
SDK compatibility without code changes
What makes it different
Bounded blast radius: compromised agents can only spend remaining epoch allowance on whitelisted targets
Passkey root stays on user's device; cloud holds only scoped, expiring session keys
Fully self-hostable open protocol with Apache-2.0 licensing
Approval workflows follow real organizational structure, not smart contract code
No token or cryptocurrency knowledge required to use the cloud platform
Three coordinated products (Relay, Loop, Lens) in one control plane versus stitching together separate tools
SDK-compatible: change the base URL, not your code
Decisions remain observable, replayable, and auditable without page turns
Full evidence bundle exports with every decision
Unified LLM gateway eliminates key sprawl and model switching without code changes