Mastyf.ai

The safety layer between your AI assistant and MCP tools

5 followers

The safety layer between your AI assistant and MCP tools

5 followers

•

•

Trust scores for MCP npm packages — plus runtime security and cost governance on every tool call. mastyf.ai scores servers with plain-English breakdowns and README badges, then enforces policy through the proxy: block injections, cap tokens per tool, throttle expensive calls, and track daily spend with burn-rate dashboards. Know what’s safe and what it costs before your agent connects.

Free

Launch tags:Developer Tools•Artificial Intelligence•GitHub

Launch Team

Viktor.comAn AI coworker that actually does the work

Promoted

Maker

📌

Hey PH 👋 — I built mastyf.ai. Too many teams were wiring AI agents to MCP servers — filesystem, GitHub, Postgres, Slack — with zero idea if those packages were safe or what the tool calls were costing them. Now you don't have to guess. The launch part: look up any npm MCP package and get a live trust score with a plain-English breakdown — CVE signals, publisher checks, category gaps, and a README badge you can embed today. Run the open-source proxy and every tool call goes through security policy and cost governance first: block risky payloads, cap tokens, throttle expensive tools, track daily spend. Founders use it to score servers before shipping vertical agents. Dev teams embed badges so clients know what's vetted. Platform folks run the proxy so production agents don't touch upstream tools ungoverned. Try it: mastyf-ai-cloud-jet.vercel.app — free cloud console, no install, no card. Source: github.com/mastyf-ai/mastyf.ai — clone and run the proxy today (npm publish coming soon). I'm here all day — tell me what you'd build or where it breaks. 🙏

Report

2mo ago