AI Dev Impact Lab — 2026-08
CLAUDE.md / .cursorrules / AGENTS.md / .windsurfrules etc.
openai / anthropic / langchain / @anthropic-ai/sdk etc. in package manifests
Estimates from random sample (N=1,698, 2026-07, ±1pt accuracy).
AI development trend
Random sample of N=2,000 from GitHub-active repos per month, classified via GraphQL. Stacked to 100%: built-only / both / using-only / non-AI (grey). AI is the colored slice — the thin bottom sliver of GitHub.
GH-Archive monthly active total (top of stack) plus sample-derived AI breakdown. Stacked: built-only / both / using-only / non-AI.
AI tools used
- Claude Code → 4,709
- AGENTS.md (Codex等) → 3,261
- Cursor → 340
- GitHub Copilot → 238
- Cline → 23
- Windsurf → 19
- Aider → 8
Featured this month
Auto-selected by AI relevance + star count. Each bar is a 0–100 score: AI = AI relevance Solo = solo dev likelihood
A trilingual (繁中 / English / 简中) learning roadmap for agentic AI: from LLM basics to multi-agent systems, with 240+ curated resources and hands-on examples. 中文 AI agent 學習地圖。
ray-r-ren/agent-apprenticeshipThe living ecosystem where AI agents complete tasks through workflow loops, improve through iterative execution, are evaluated by mentor agents or humans in the loop, and turn completed work into reusable work experience and data to improve future agents.
LiteLLM-Labs/litellm-agent-control-plane1 place to call all your agents - OpenCode, Hermes, Claude Managed Agents, Cursor Agents API, DeepAgents.
fitchmultz/pi-cursor-sdkRun Cursor's agent loop inside the pi coding agent via local Cursor SDK agents, with native model selection, thinking controls, fast/plan modes, image input, and MCP-bridged pi tools.