HAPI MCP
Headless API stack that turns existing OpenAPI specs into MCP tools for AI agents without rewrites.
At a Glance
About HAPI MCP
HAPI MCP, published by La Rebelion Labs under MCP.com.ai, converts existing OpenAPI specs into Model Context Protocol (MCP) tools so AI agents can call APIs directly. The site describes a stack of components: HAPI Server, runMCP, OrcA, QBot and chatMCP.
What It Is
HAPI MCP is a "Headless API" approach to MCP. A CLI takes an OpenAPI/Swagger spec and generates an MCP server, so the API remains the runtime and single source of truth rather than being re-implemented in agent code. The vendor states that auth, RBAC, rate limits and audit trails are inherited from the underlying API.
How the Workflow Runs
The site describes three steps: pick an API spec (OpenAPI 3.0+, REST), run a single CLI command, and use the resulting MCP server as a tool, agent backend or testing interface. The install is a shell script fetched via curl. Updating the spec updates the tools.
The Component Stack
- HAPI Server: generates MCP tools from OpenAPI.
- runMCP: autoscaling execution and testing of MCP tools.
- OrcA: deterministic orchestration of multi-tool tasks using OpenAPI/Arazzo; also available as a VS Code extension acting as an MCP agent harness with progressive tool discovery, Ask/Plan/Act roles and a traffic analyzer showing each LLM round and MCP tool call.
- QBot: terminal TUI client for MCP tools.
- chatMCP: conversational MCP-native client.
Client Compatibility
The vendor states any MCP client can consume the tools, including ChatGPT, Claude, QBot, Agentico.dev and custom agents. Deployment options listed include on-premise and Cloudflare.
Community Discussions
Be the first to start a conversation about HAPI MCP
Share your experience with HAPI MCP, ask questions, or help others learn from your insights.
Pricing
Free
Explore HAPI locally or in a low-volume environment; no registration required.
- 36,000 MCP executions / month
- 20 per minute
- npm/Bun packages, native binaries, and Docker
- Community support
- On-prem or fully managed on the cloud
Starter
For a first production integration and small engineering teams.
- 250,000 MCP executions / month
- 120 per minute
- Commercial production capacity
- Selected production capabilities
- Standard support
- Fully managed on the cloud
- Production plugins (Open Telemetry, OAuth2, Webhooks, and more)
Pro
For teams making API capabilities part of their product.
- 2 million MCP executions / month
- 600 per minute
- Production capability set
- Priority support
- Capacity for sustained workloads
- Fully managed on the cloud
- Production plugins and Add-ons
Capabilities
Key Features
- Generate MCP servers from OpenAPI/Swagger specs via CLI
- Inherits API auth, RBAC, rate limits and audit trails
- runMCP autoscaling execution and testing
- OrcA deterministic orchestration with OpenAPI/Arazzo
- OrcA VS Code agent harness with progressive tool discovery
- Ask/Plan/Act agent roles
- MCP traffic analyzer in VS Code
- QBot terminal client
- chatMCP conversational client
- OAuth 2.0 Dynamic Client Registration support