AI agents are entering your codebase.
Fleet makes sure your security team is OK with that.
Data plane on your infrastructure; your source code stays private. Full audit trail on every agent decision. Full run tracking on every agent.
Local-first execution with explicit network boundaries
Know what runs locally, what leaves the worker, and which services receive it.
Single binary
One Go worker binary with no Node.js or Docker runtime on the host.
A small runtime surface with one Fleet artifact to verify and update
Your code stays private
Agents run on your machines. Source code goes only to your configured model backend and GitHub; operational metadata, run usage, and synced audit events reach the dashboard.
The Fleet dashboard does not receive your source code
Embedded worker state
Local worker state lives in embedded SQLite; Fleet hosts the control-plane storage used by the dashboard.
No customer-managed database is required for the worker
Local Unix-socket IPC
Local daemon coordination uses Unix domain sockets, so the worker does not need an inbound application port by default.
No default inbound Fleet service to expose
Data sovereignty
Claude Code's model backend is configurable — use the Anthropic API directly, or point it at Amazon Bedrock / Google Vertex to keep model traffic in your own cloud. Your source code stays private; the dashboard receives operational metadata + usage metering, never code.
Keep model traffic in your cloud when Bedrock or Vertex is configured
No JS/Docker runtime
No Node.js, no Docker daemon, no container orchestration needed.
Minimal attack surface, minimal maintenance
Compliance frameworks
Designed to support regulatory requirements out of the box with built-in governance and audit capabilities.
Data sovereignty
ConfigurableClaude Code can use Bedrock or Vertex in your own cloud. Your source code stays private; the dashboard receives operational metadata and usage metering, never code.
Audit trail
Built-inDesigned to support SOC 2, ISO 27001, and internal audit logging requirements
Run tracking
Built-inPer-agent cumulative run time, session counts, and time-based limits
Agent quarantine
Built-inAutomatic isolation of agents exhibiting anomalous behavior
Approval gates
Built-inHuman-in-the-loop controls at any stage of the agent workflow
Role-based access
Built-inOrg hierarchy with department and team-level scoping
Deployment options
Run the worker on your infrastructure and choose the model backend and connected services that fit your boundary.
Data plane on your infrastructure
The Fleet binary and agents run on your machines. Dashboard and analytics at app.fleetctl.ai.
Restricted egress
Allow the services you configure: your model backend, GitHub, app.fleetctl.ai, PostHog's anonymous CLI telemetry endpoint unless you opt out, and any connected Linear or Jira provider.
Managed
Let us handle infrastructure, updates, and monitoring. Focus on building, not operating. Contact us for availability.
Enterprise governance
Full organizational control over your AI fleet with hierarchical oversight and policy enforcement.
- Org hierarchy with CEO, CPO, CTO, department heads, and repo-scoped developers
- Role-based access control across org and repo tiers
- Per-agent run tracking with cumulative time and session counts
- Complete audit trail via unified fleet log with decision history
- Pipeline approval gates with rejection routing and retry logic
- Agent quarantine for isolating misbehaving agents instantly
Deep observability
Continuous monitoring, evaluation, and risk assessment across your entire fleet.
6-dimension eval scoring
Task output, reliability, quality, efficiency, collaboration, and cost scoring on every agent run.
Risk model
8-feature risk model predicts agent failures before they happen.
Unified fleet log
Chronological timeline merging events and activity across all repos.
Anomaly detection
Real-time analysis of agent behavior for automatic quarantine.
Enterprise support
White-glove support from engineers who understand fleet orchestration at scale.
Dedicated engineer
A named support engineer who knows your deployment inside and out.
SSO via Auth0
Single sign-on for the Fleet dashboard via Auth0. Supports SAML, OIDC, and social providers.
Service terms
Discuss dashboard uptime, response channels, and support commitments during enterprise review.
Training
Onboarding workshops and ongoing enablement available on request.
Architecture review
Architecture reviews to optimize your fleet topology and workflows.
Volume pricing
Custom per-fleet and volume pricing for large deployments. Contact us for details.
Licensing & procurement
The binary is free to install and inspect; starting agents requires a license. Team is $299/fleet/month with unlimited roles; enterprise terms, invoicing, and security review support via this page's contact form.
Talk to our team
Tell us about your fleet requirements and we'll get back to you within one business day.
No automated follow-up sequences.