agent-lsp: agent-lsp is a stateful runtime over real language servers. It indexes your workspace once, keeps the index warm, and adds a skill layer that encodes correct multi-step operations. It provides speculative execution, phase enforcement, and token savings for AI agents.; fastmcp: FastMCP is a standard framework for building Model Context Protocol (MCP) applications, which connect LLMs to tools and data. It simplifies the process by automatically generating schemas, validation, and documentation for tools, and managing transport negotiation and authentication for server connections. FastMCP offers a comprehensive solution for developing, deploying, and scaling MCP-powered systems.
Multi-project sessions: work across projects without reconfiguring
Building LLM applications that interact with custom tools and data sources