prime-radiant-inc

Claude Plugin Stats — Development skill for Claude Code

Development community

Daily scrape of Claude Code plugin install stats.

How to install Claude Plugin Stats

This entry records only its repository, not the path inside it, so there is no exact command to give. Open prime-radiant-inc/claude-plugin-stats and copy the folder into ~/.claude/skills/, or the file into ~/.claude/agents/.

What Claude Plugin Stats does

Daily scrape of Claude Code plugin install stats.

Alternatives in Development

Codexbar — Show usage stats for OpenAI Codex and Claude Code, without having to login 8.8k ★
Scrape Jobs — List or clean up Google Maps scrape jobs 1.4k ★
Sniffly — Claude Code dashboard with usage stats, error analysis, and sharable feature 1.2k ★

README

claude-plugin-stats

Daily scrape of Claude Code plugin install counts from Anthropic's public plugin-stats endpoint, with history back to **December 2025** recovered from Arq backups.

📈 [Interactive chart of installs over time →](https://primeradiant.com/claude-plugin-stats/)

[![Plugin installs over time](docs/preview.png)](https://primeradiant.com/claude-plugin-stats/)

Search any of the 450+ plugins, toggle log scale, zoom into date ranges, and hover for exact counts. (GitHub strips JavaScript from READMEs, so the interactive version lives on GitHub Pages — click through.)

How it works

A GitHub Action fetches `https://storage.googleapis.com/claude-code-dist-86c565f3-f756-42ad-8dfa-d59b1c096819/plugin-stats/plugin-details.json` daily and commits it if it changed — [Simon Willison's git-scraper pattern](https://simonwillison.net/2020/Oct/9/git-scraping/). The same run regenerates `docs/data.json` (one `[date, installs]` series per plugin, extracted from the full git history) which feeds the chart.

The recovered history

The scrape only started in June 2026, but Claude Code has cached this data locally since late December 2025 (`~/.claude/plugins/install-counts-cache.json`, later renamed `plugin-catalog-cache.json`). Weekly snapshots of those files were recovered from encrypted Arq 7 backups in S3 — including thawing the relevant pack files out of Glacier — and committed with their original `fetchedAt` timestamps as commit dates. The patched restore tooling that made this possible lives at [obra/arq_restore](https://github.com/obra/arq_restore); the recovery process is documented in [RECOVERY-HANDOFF.md](RECOVERY-HANDOFF.md).

Files

plugin-details.json — latest snapshot from the public endpoint
install-counts-cache.json, plugin-catalog-cache.json — historical local cache formats (frozen; present for the recovered history in git log)
scripts/build-history.py — walks git history → docs/data.json
docs/ — the GitHub Pages chart