AI Dev Impact Lab JA
← Monthly reports · REPORT 2026-06 · published 2026-07-02
MONTHLY REPORT

June 2026 AI Development Impact Report

Observed facts: the composite AI activity score for 2026-06 is 337.05 points. The reported month-over-month delta is -75.97 points and the year-over-year delt

851
AI repos
248
tool signals
94
solo dev
105
web launches
COMPOSITE SCORE
337.05
MoM -76.0%
vs 2021 +237.1%
12-month trend
🥇 TOP TOOL
Claude Code
145
repos
60.7%
share

ADOPTION AI coding tool adoption — June 2026

CONTEXT Global GitHub scale (GH Archive)

5,039,271
new repos
10,040,434
create events
95,398,281
push events
68,905
forks
310,744
watch events

AI-relevant share: 0.0169% (851 / 5,039,271) · with tool signals: 29.1%

ANALYSIS Analysis & commentary

Executive summary

Observed facts: the composite AI activity score for 2026-06 is 337.05 points. The reported month-over-month delta is -75.97 points and the year-over-year delta is +215.74 points. There were 851 AI-related repos (851 repos) and 248 repos with detectable AI-tool signals (248 repos). Continuity rate is reported as 0 (confidence: low).

Interpretation: observed values show elevated activity compared with the longer baseline (yearly increase) but a pullback relative to the immediate prior month. The low-confidence continuity metric limits claims about ongoing maintenance or developer follow-through for newly created projects.

Key movements

Observed facts: the composite score moved from 1,402.45 points in 2026-05 to 337.05 points in 2026-06. Top AI tools this month were led by claude_md with 145 repos and agents_md with 75 repos. Solo-development candidates numbered 94 repos and web-launch candidates numbered 105 repos.

Interpretation: the sharp reduction in composite score from May suggests a concentration of activity in the prior month or a measurement sensitivity; tool signals remain concentrated around a small number of tool markers. The counts of solo and web-launch candidates indicate continued individual contributor experimentation, but not large-scale public launches.

Global context

Observed facts: platform-wide totals include 11,753,398 active repos and 95,398,281 push events. Total public repos created stands at 5,039,271 repos and total create events are 10,040,434 events. Watch events are 310,744 events.

Interpretation: AI-related signals represent a small fraction of overall platform activity by raw counts. The absolute platform scale implies that changes in AI-specific metrics can reflect either genuine shifts in developer focus or sampling/annotation variance in signal detection.

Outlook

Observed facts: leading repositories this month include Forsy-AI/agent-apprenticeship (595 stars, AI score 100) and several smaller projects with AI scores ranging up to 100. The composite score recent history shows a long period near 106.75 points through 2026-04, a spike to 1,402.45 points in 2026-05, and 337.05 points in 2026-06.

Interpretation and implications: expect continued episodic bursts of visible AI activity driven by a few high-impact projects or tool integrations. Given the low confidence on continuity, monitor push and commit patterns and tool-signal coverage in subsequent months before assuming sustained development or community adoption.

REPOS Top repos in June 2026 (AI score 40+ / by stars)

Forsy-AI/agent-apprenticeship

The living ecosystem where AI agents learn from real-world work through iterative workflow loops, reusable experience, and collective training signal exchange.

595 AI 100
gameworkerkim/cassandra-ai
TypeScript 41 AI 50
IssacW228/student-llm-wiki

📚 Student LLM Wiki — AI-compiled knowledge base for university students. Drop course slides, get a persistent interlinked wiki. Feynman review, exam prep, confidence decay, cross-course connections. Based on Karpathy's LLM Wiki pattern. Works with Claude Cowork / Claude Code / Claudian + Obsidian.学生大语言模型维基百科——专为大学生打造的AI编译知识库。

33 AI 100 3 sig
dujunyi416/claude-nightshift

Quota-aware automation for Claude Code: check rate limits, pre-warm the 5h window before you wake, queue prompts overnight, auto-resume interrupted sessions on reset.

Python 26 AI 75
richchang0721-boop/it-diagnostic-agent

互動式 IT 故障排除 Runbook,整合 Claude AI。涵蓋 CCTV、NAS、AD、Exchange、Outlook 等企業常見場景。純靜態 HTML,GitHub Pages 一鍵部署。AI-powered IT troubleshooting runbook for Active Directory, Exchange, Outlook, NAS, CCTV and manufacturing environments. Designed for real-world IT operations and support teams.

HTML 18 AI 90
nnguyenquangg/claude-code-spec-driven-development
Shell 18 AI 50
SuperJayLiu/AI-for-Economics-and-Finance-Research

An "all-you-need-to-know" practical guide to using AI and LLMs responsibly and effectively for economics and finance research: from core skills and ChatGPT/Claude Projects to coding, writing, reproducibility, and automation. 小红书账号:SuperJay

HTML 14 AI 100
RiggdAI/uniqent

Any brain, any agent. Package an AI agent's brain — persona, MCP stack, skills, memory, config — into one open, signed bundle and install it into any framework (OpenClaw, Hermes, Claude Code) in one click.

TypeScript 14 AI 70 1 sig
su5176/Mklink-AI-Probe

MKLink AI Probe is an all-in-one embedded debugging CLI for Cortex-M microcontrollers. It bridges the MKLink/MicroLink hardware probe with AI agents (Claude, OpenAI) to enable natural-language driven firmware flashing, real-time waveform visualization, memory inspection, fault diagnosis, and industrial protocol debugging.

Python 11 AI 65
KorroAi/korrodesign

Design Copilot for Claude Code — two enforcement layers (Taste Guardian + Blind Spot ESLint plugin). Works standalone. Zero dependencies. Awwwards-level output.

JavaScript 10 AI 45
mythkiven/rednote-director-skill

面向小红书/RedNote 创作者的 Agent Skill:图文轮播规划、视觉导演、封面内页设计、AI 生图提示词与发布文案。支持 Cursor、Claude、Codex。

9 AI 70 1 sig
hoangpm96/ai-switcher

A native macOS app to manage and switch between multiple accounts for AI coding tools — Claude Code, Codex, and Antigravity IDE — from one place.

Rust 7 AI 75 1 sig

NEW New AI repos created in June 2026

Forsy-AI/agent-apprenticeship

The living ecosystem where AI agents learn from real-world work through iterative workflow loops, reusable experience, and collective training signal exchange.

595 AI 100 NEW
IssacW228/student-llm-wiki

📚 Student LLM Wiki — AI-compiled knowledge base for university students. Drop course slides, get a persistent interlinked wiki. Feynman review, exam prep, confidence decay, cross-course connections. Based on Karpathy's LLM Wiki pattern. Works with Claude Cowork / Claude Code / Claudian + Obsidian.学生大语言模型维基百科——专为大学生打造的AI编译知识库。

33 AI 100 NEW
chu459/claude-code-orchestrator-skill

World-class Codex Skill and MCP harness for orchestrating Claude Code workers through CCSwitch multi-model routing.

Python 2 AI 100 NEW
doivamong/agent-workbench

A practical workbench of tools + methodology for solo devs driving an AI coding agent (Claude Code, Cursor, Copilot): safety hooks, secrets crypto, invariant + test-selection tooling, leak scanner, plus skill & memory system templates.

Python 2 AI 100 NEW
boggspa/TaskWraith

Local-first macOS workbench for running and reviewing AI coding agents — multi-provider, with multi-agent ensembles, diff review, and on-device history.

TypeScript 1 AI 100 NEW
drobertson-dev/agent-studio

Self-hosted AI application studio with a Nuxt UI, agent runtime, live sites, runtime API routes, and Docker Compose deployment.

Vue 1 AI 100 NEW
Junaid-Ahmed-Rupok/continuum-rag-chatbot

🧠 A persistent memory RAG chatbot that never forgets. Uses Phi-3-mini LLM, ChromaDB vector database, and Ebbinghaus memory decay curves.

Python 1 AI 100 NEW
jackcongmac/claude-codex-bridge

Bidirectional MCP bridge between Claude Code and Codex — each agent calls the other as a tool, with the Claude side as a persistent, project-aware colleague.

Python 1 AI 100 NEW

EXPLORE Dig deeper

📅

June 2026 raw snapshot

Top 20 repos, all monthly metrics, GH Archive globals

⌨️

AI coding tool ranking

Claude Code, Cursor, Codex, Copilot, Windsurf, Aider, Cline

📊

All 13 rankings

languages, LLMs, frameworks, DBs, models, embeddings, agents

← Previous May 2026 AI Development Impact Report

Aggregated from snapshots of AI-relevant repos (AI score ≥ 40) for this month. methodology