Parkour Mcp

Parkour Mcp is a Model Context Protocol (MCP) server that enables AI assistants and LLMs to surface high-signal, unsummarized web content through structured API integrations. It provides tools for targeted content extraction, citation tracking, and knowledge synthesis, leveraging clean APIs (Kagi, Semantic Scholar, arXiv, GitHub, MediaWiki, Reddit, etc.) and advanced Markdown conversion. With features such as intelligent frontmatter (YAML) envelopes, cache for faster followup queries, and responsible citation nudges, it is particularly useful for AI agents, researchers, and developers integrating live, citation-friendly web content into their workflows.