Arcade
Actions runtime that authorizes, executes and audits agent tool calls across enterprise systems
Arcade is an actions runtime that sits between developers' AI agents and the systems they act on. It authenticates users, lets agents act on their behalf within delegated scopes, executes tool calls across business systems and logs every action from one control plane.
Key features:
- Call more than 8,000 permission-aware tools or bring custom MCP servers.
- Authenticate through the customer's identity provider with delegated, per-user authorization.
- Enforce security policies outside the model, where a prompt cannot undo them.
- Log every action for audit and SIEM integration.
- Reduce token use with tool recommendation and test tools with built-in evals.
- Use it from Claude Code, Cursor, Codex, VS Code, Copilot CLI and Claude Desktop.
Arcade runs on Arcade Cloud, with VPC and air-gapped deployment on Enterprise. The free plan includes 2,000 auth events and 2,000 tool calls a month; Team is $25/month plus $0.10 per auth event and $0.01 per tool call, and Enterprise adds SSO, RBAC and audit logs.