Skip to content

pingcap/ossinsight

Repository files navigation

OSSInsight

The analytics engine for the AI-native open source ecosystem.
Analyze 10+ billion GitHub events. Track AI agents, coding tools, and the repos shaping the future.

What is OSSInsight?

OSSInsight analyzes 10+ billion rows of GitHub event data to surface insights about the open source ecosystem — from individual developers to entire technical fields.

In 2026, that means tracking the explosion of AI agents, coding assistants, research automation, and the infrastructure being built around them. OSSInsight is how you see what's actually happening in open source, measured in commits, stars, forks, and contributors — not hype.

For AI builders

  • AI Agent Frameworks — Rankings and trends across LangChain, CrewAI, AutoGen, and 50+ agent frameworks
  • Coding Agents — Track Claude Code, Copilot, Cursor, Aider, and the autonomous coding wave
  • Research Agents — Analyze repos like autoresearch (54K stars in 19 days) that are turning research into search
  • MCP & Tool Infrastructure — The standardizing integration layer for AI agents

For developers

For researchers & analysts

  • Data Explorer — Ask questions about GitHub data in natural language, get SQL + visualizations
  • 60+ Curated Collections — From databases to Web3, from DevOps to AI safety
  • Blog — Data-driven analysis of open source trends

LLM-Friendly

OSSInsight is built for the AI era:

  • /llms.txt — Structured site description for LLMs
  • /llms-full.txt — Full documentation in LLM-friendly format
  • OpenSearch — Machine-readable search integration
  • Schema.org structured data on every page — TechArticle, CollectionPage, BreadcrumbList, FAQPage, and more

Featured Analysis

Topic What we found
autoresearch: 54K Stars in 19 Days Research is becoming search. karpathy/autoresearch has a 1,085:1 fork-to-contributor ratio — people fork to run private experiments, not to contribute back.
The Coding Agent Wars Claude Code, Codex, OpenCode — the autonomous coding landscape mapped by the data.
Agent Skills: Not the Endgame 57K AGENTS.md repos, 21K CLAUDE.md — skills are a transitional layer, not the final form.

Features

Data Explorer

Ask questions about GitHub data in natural language — Data Explorer generates SQL, queries the data, and presents results visually.

Examples:

Collections & Rankings

Find insights about monthly or historical rankings and trends in technical fields with curated repository lists.

GitHub Collections Analytics

Examples:

Developer Analytics

Insights about developer productivity, work cadence, and collaboration from contribution behavior.

  • Contribution time distribution, stars, languages, and trends
  • Code (commits, pull requests, code line changes), code reviews, and issues
Developer Analytics

Repository Analytics

Insights about code update frequency & popularity from repository status.

  • Stars, forks, issues, commits, pull requests, contributors, languages, and lines of code
  • Geographical and company distribution of stargazers, issue creators, and PR creators
Repository Analytics

Examples:

Compare Projects

Compare two projects side-by-side on any metric.

Examples:

Collections

Curated lists of repos in technical fields, ranked by GitHub metrics. Perfect for tracking ecosystems.

Add a collection by submitting a PR to etl/meta/collections/:

id: <collection_id>
name: <collection_name>
items:
  - owner/repo-1
  - owner/repo-2

Popular collections: AI Agent FrameworksOpen Source DatabaseWeb FrameworkJavaScript ORMMore...

Development

pnpm install
pnpm dev        # Start web app (port 3001)
pnpm dev:docs   # Start docs site (port 3002)
pnpm dev:all    # Start both

Contributing

Contact

Powered by
Image

About

Analysis, Comparison, Trends, Rankings of Open Source Software, you can also get insight from more than 10 billion with natural language (powered by LLM). Follow us on Twitter: https://twitter.com/ossinsight

Topics

Resources

License

Stars

Watchers

Forks

Contributors