AI & MCP

MCP mocks, authenticators and test environments for agents.

13 tools13 available now

Test walletGive agents fake USD test balance, spending limits and human approval for test payments.Checking availability

Test x402 paywallTest HTTP 402 payment flows with wallet tokens or permissive fixtures. No real settlement.Checking availability

Test checkoutExercise checkout outcomes and signed webhooks. Test mode, no real payment.Checking availability

Password vaultStore encrypted credentials for accounts your human asked you to create or operate, with approval for password reveals.Checking availability

Agent authenticatorStore encrypted TOTP authenticators and get sign-in codes for accounts your human asked you to create or operate.Checking availability

Mock MCP serverGive agents a temporary MCP server with configurable tools.Available

MCP tool-call inspectorInspect tool calls, responses, and request history.Available

MCP server inspectorExplore the capabilities exposed by a remote MCP server.Available

Mock LLM endpointTest model integrations against configurable responses.Available

Mock agents & LLMsScript agent-facing model text using OpenAI Chat Completions or Anthropic Messages response shapes.Available

LLM streaming simulatorTest paced assistant text and streams that end before the completion event.Available

Tool failure simulatorTest how an agent handles errors and tool latency.Available

Kivel MCPCreate and manage platform resources from your AI agent.Available

Full category roadmap

All the capabilities planned for this category. Availability is shown on each tool above.

Mock MCP servers and tool-call history
Scripted OpenAI and Anthropic text responses
Paced and interrupted SSE streams
Workspace memory and human approvals

“Coming soon” means planned, not available. We’ll only mark a tool ready when you can actually use it.