Capacity and published usage terms.
The published plan measures connected servers and independently enrolled managed workloads, labeled managed services in the plan. One enrollment can contain an API and worker in its manifest and counts as one unit; a separately enrolled workload uses another unit. Child processes within one workload are not individually billed under the current workload-capacity model. The plan includes two connected servers and five managed services. Additional connected servers are $19/month each, and additional managed services are $9/month each.
How you get up and running
The supported path starts with account access and the signed Trust Console, then uses infraveil.json to connect and configure customer-managed services.
Create an account and complete the portal steps available for your workspace.
Use the signed Trust Console to connect a customer-managed server.
Configure a service definition and review the available operational records.
Configure supported policy and monitoring for managed routes as needed.
Professional
For configured backend services on customer-managed servers
Published monthly rate. No setup fee.
- Published Professional plan access
- Five managed services included
- Two connected servers included
- Additional connected servers: $19/month each; additional managed services: $9/month each
- Configured runtime health reporting
- Configured gateway policy for managed routes
- No published usage metering for traffic, logs, requests, or security events
- WebAuthn + TOTP authentication
- Features subject to current documentation and workspace configuration
- Signed agreements control any different commercial terms
Questions about the published plan
The rates and included capacity above are the public terms. A signed agreement controls only when it expressly states different commercial terms.
Configured capabilities
The published plan describes capacity and usage-metering terms. Product capabilities depend on supported configuration, the current documentation, and any signed agreement.
Health Monitoring
Configured launcher and agent communications can report runtime health and operational metrics.
Recovery Workflows
Configured recovery can use local cache and release state. It remains bounded by host state, connectivity, and customer configuration.
Built-in Security
For managed gateway traffic, configured policy can include rate limits, honeypot routes, and suspicious-pattern detection.
Centralized Logging
App Logs show captured application output; Runtime Logs show supervisor and control activity. Request and Control Trace describe their respective observed paths. Queues and retention are bounded, and missing or dropped observations remain visible limitations.
WebAuthn + TOTP
Biometric sign-in (Windows Hello, Touch ID) and authenticator app two-factor for dashboard access.
Managed Runtime Storage
Legacy Python payloads may run from memory. Manifest-defined and supervised services use a managed workspace and cache on the customer host for execution and recovery.
Frequently Asked Questions
How do assistant integrations work?
Connect a supported MCP client through the workspace’s scoped, expiring connection. Named tools cover supported inspection and operating actions; changes follow authorization and preview/apply checks. The client receives neither unrestricted shell access nor the customer release-signing key.
What does the Pipeline Builder do?
Configured pipelines apply supported telemetry and response stages to managed traffic. Their declarative rules and bounded analysis hooks do not execute arbitrary code or give an external model authority to rewrite policy.
Is there another public plan?
Professional is the only published platform plan. It includes two connected servers and five managed services. Additional connected servers are $19/month each, and additional managed services are $9/month each.