HasMCP

HasMCP is a MCP Framework that creates remote MCP Server with Streamable HTTP from API definitions.

Automated OpenAPI Mapping HasMCP bridges existing APIs and Large Language Models (LLMs) by automatically translating OpenAPI (v3.0/3.1) or Swagger files into the Model Context Protocol (MCP). This removes the need for developers to write "glue code," allowing LLMs to interpret APIs as executable tools instantly. This ensures high accuracy and a no-code integration experience.

Native MCP Elicitation Auth To solve authentication security, HasMCP implements "Elicitation Auth" for OAuth2. Rather than exposing credentials to an LLM, the system pauses to provide a secure login URL to the user. Once authenticated via standard providers (e.g., Google), HasMCP manages the secure token, ensuring user credentials are never stored insecurely or accessible to the AI.

Context Window Optimization This feature reduces cost and latency by minimizing the data sent to the LLM. Since LLMs have limited "memory" (context windows) and charge per token, HasMCP filters irrelevant API response data. This ensures faster processing and helps the LLM focus only on relevant information. Optimization is achieved via two primary methods:

JMESPath Pruning: A declarative query language used to filter JSON. It strips away unused fields (noise) from API responses, keeping only what is requested.

Goja (JS) Logic: An embedded JavaScript engine that handles complex data manipulation. It allows for procedural logic, data combination, and formatting before the data reaches the LLM.

Real-time Dynamic Tooling HasMCP ensures LLMs always have an up-to-date list of tools without requiring server restarts. By monitoring API health and user permissions, the system triggers tool_changed events. If an API goes down or permissions change, the LLM allows for immediate adaptation, ensuring resilience and security.

Secure Secret & Proxy Management This feature centralizes security by storing sensitive data (like API keys) in an encrypted vault. HasMCP acts as a proxy, injecting these secrets into requests on the fly. This prevents hard-coding secrets and allows for flexible header management, ensuring sensitive keys are never exposed to the LLM or end-users.

MCP Composition Focused on modularity and performance, this allows developers to chain multiple MCP servers together. By composing dependencies (e.g., combining a weather server with a restaurant server), users can build scalable, complex, and high-performance AI agents from reusable components.

Observability & Telemetry HasMCP provides a suite of tools to monitor and debug AI agents. Features include:

Tool Call Analytics: Tracks most-used tools.

User Governance: Audits usage per user.

Token Economics: Quantifies cost savings from context optimization.

Streaming Debug Console: Real-time event logs for debugging.

Payload Inspector: Visualizes data before and after transformation.