maker here. I built this because I was doing SEO and ASO work across four properties and kept hitting the same two problems. the first is that the consoles do not talk to each other. Search Console, App Store Connect, Play and wp.org each have their own dashboard, their own auth, their own idea of what a week means. answering "how did we do" meant four tabs and a lot of retyping. the second is worse and it is why snapshot exists. nobody can diff a dashboard against what it said three weeks ago. App Store Connect keeps no history of a subtitle once you overwrite it. so "did that change work" gets answered from memory, and memory is how something that got worse gets remembered as flat. snapshot writes every surface to one timestamped file, and compare_snapshots does the arithmetic between two. the CLI matters for the same reason. over MCP a result lands in a model's context and is gone. from the shell it lands in a file you can still read in September. a few things I learned building it that might save someone else the time. Apple indexes the app name, the subtitle and the hidden keyword field, and does not index the description at all, so tools that measure description length are measuring nothing. Apple's API has no aggregate star rating resource, only age ratings, so any rating you see came from the public iTunes endpoint and should say so. Google Play's bulk report CSVs are UTF-16 with a byte order mark. and Search Console silently truncates at rowLimit, which nearly had me conclude a product had no demand when the row I needed was one past the cut, so the tool now fetches rowLimit+1 and tells you when there is more. MIT, 28 tools, no key needed for the on-page audit, PageSpeed, IndexNow, keyword ideas or wp.org tools. happy to add a surface if you are measuring something it misses.

seo-console-mcp

One MCP server for Search Console, the App Store and Play

One MCP server for Search Console, the App Store and Play