Home

Documentation

What is Maybe Don’t?

Maybe Don’t provides guardrails and observability for agentic AI. It sits between AI agents and the tools they use — MCP servers and CLI commands — evaluating every operation against your policies before it executes. You define the rules, Maybe Don’t enforces them.

Think of it as a security checkpoint for your AI’s actions. Every tool call and CLI command is audited, validated, and logged.

ClaudeChatGPTCursorGeminiAI AgentsMaybe Don'tMCP GatewayPolicy EngineAudit LogGitHubSalesforceLinearCloudflareDownstream MCP Servers

Why Use It?

  • Audit everything — See exactly what your AI agents are doing, across MCP tool calls and CLI commands
  • Set guardrails — Block dangerous operations before they happen, using AI or deterministic policies
  • Stay in control — Enforce policies, review decisions, and trace any action back to its source
  • Observe and learn — Run in audit-only mode to understand agent behavior before enforcing rules

Get Started

Get Started
Up and running in minutes
MCP Gateway
Validate MCP tool calls
CLI Gateway
Validate CLI commands
Configuration
Global settings, logging, and reference
Policies
Define what’s allowed and what isn’t
Audit Log
The record of everything
Testing
Test your policies before deploying
Cookie PolicyLicense