Endara is a local-first MCP relay designed to provide AI clients with unified access to both cloud services and local resources through a single endpoint on a user's machine. By connecting to the relay at localhost, AI applications can interact with tools such as Gmail, GitHub, Slack, Postgres, and local files without the need to configure each client individually for every service. This approach enables seamless integration of multiple services, whether cloud-based or local, under one connection.
The platform offers a native desktop application for macOS, Windows, and Linux, allowing users to manage endpoints, monitor health, inspect tools and schemas, and view real-time logs through a graphical interface. Users can add or remove MCP servers, browse a curated marketplace of available servers, and configure multiple accounts for the same service, each with its own description visible to the AI model. The tool supports hot-reloading of configuration changes without requiring a restart. Endara also features a system tray presence and background operation for convenience.
1 with PKCE, supporting sign-in and token management for services like Linear, Notion, Slack, GitHub, and Sentry. It handles token refresh, lifecycle management, and automatic reconnection, enabling AI clients to securely access cloud tools. The platform is compatible with any MCP client and supports various transports including stdio, SSE, HTTP, and OAuth. Tools are auto-prefixed to avoid naming collisions, and users can install the same server multiple times for different accounts or projects.
A notable feature is the JavaScript execution mode, which condenses a large catalog of tools into three meta-tools—search, list, and execute—allowing the AI model to orchestrate multi-tool workflows by generating and running scripts in a sandboxed environment. This reduces complexity for the model and enhances orchestration capabilities. 0 license, and its source code is available for review and contribution. The tool classifies as a framework or SDK for integrating AI clients with diverse resources via the MCP protocol.
Endara sits in PulseGate's Frameworks & SDKs category. It focuses on connecting AI agents to both cloud services and local resources through a unified local endpoint. Endara is an open-source project aimed at AI developers. The project is open source (Apache-2.0). Endara is available on the command line, macOS, Windows, Linux, and API, and it can be self-hosted.
endara-ai builds and maintains Endara, and the product first shipped in 2026. The project is developed in the open on GitHub with 23 stars and 126 commits in the last 90 days. Across PulseGate's embedding index, Endara has few near neighbours, marking it as relatively distinct. Among its 6 catalogued features are MCP relay, local-first, and API endpoint. It exposes integrations via an MCP server and a public API.
Latest indexed changes and source events
Other apps tracked under the same category.