Show HN: FreshContext – exponential decay scoring for retrieved web data`

Category: ai-ml

Tags: mcp-server, ai-agents, data-freshness, web-scraping, typescript

Score: 7.3/10 (Innovation: 7, Technical: 6, Documentation: 8, Utility: 8)

FreshContext is an MCP server and open standard that adds timestamp and confidence metadata to every piece of web data an AI agent retrieves, enabling freshness-aware reasoning. It packages 20+ tools for scraping diverse sources (GitHub, HN, SEC filings, etc.) into parallel landscape queries, solving the common problem of AI tools presenting stale or dead links as fact. Its novelty lies in making retrieval time a first-class citizen for AI agents, with a structured envelope and decay scoring system.

Target audience: backend devs, data engineers, AI engineers

Repository: https://github.com/PrinceGabriel-lgtm/freshcontext-mcp · TypeScript · 6 stars

View on Hacker News