SteamMirror

Steam knows a lot about how you play, but it shows you almost none of it. SteamMirror takes a public Steam ID or profile name and turns it into a picture of your gaming life: where your hours actually go, which games you finished, what's been sitting in your backlog for years, and what's on your wishlist.

What it does

• Playtime breakdown: your library sorted by where the time really went, with charts for recent and all-time play.
• Player scores: formula-based scores for intensity, exploration, commitment, mastery, and backlog. No AI guesswork; the same library always gets the same result.
• Achievements and wishlist: completion rates and what you're waiting to buy, side by side with what you already own.
• Shareable profiles: every analysis has its own URL, and you can export your data as CSV.
• Open in LLM: one click sends a ready-made prompt to the AI assistant of your choice for game recommendations. SteamMirror never stores an AI key or pays for AI calls.

How it works

Everything runs on Cloudflare's edge. A Worker calls the Steam Web API, computes the scores, and caches results in D1 and the edge cache so popular profiles load instantly and Steam isn't hammered. Profiles are stored only as SHA-256 hashes, and each IP is rate limited.

The global analytics page shows which games the community plays most. Ranking by total hours turned out to be misleading: a couple of players with 10,000 hours in one game could take over the whole list. Now every profile gets one vote, split across its games by share of playtime. That way a game ranks high when many people spend a large part of their time on it, not when one person never stops playing it.