Welcome back! There are 26 slash commands hiding in Claude Code and most people use three. We also found seven Adobe replacements written in Rust, the GitHub repos worth your time, and a clean way to decide between RAG, CAG and KAG.
In today's Generative AI Academy Newsletter:
26 Claude Code commands most users don’t know exist: Start with 13, including one that asks a side question without polluting your conversation.
One developer just made Adobe optional: Seven apps, clean-room, native Rust, open source and downloadable today.
12 GitHub repositories every AI developer should know: From a 392K-star local assistant to the agent framework most teams end up using.
RAG, CAG and KAG solve the same problem in 3 very different ways: Docs change daily, pick one. Small and stable, pick another.
26 Claude Code commands most users don't know exist
Type them at the start of a message in Claude Code (not the regular Claude chat app).
The 13 we’d learn first:
→ /btw asks a side question without adding it to your conversation
→ /context shows what’s eating your context window
→ /goal keeps Claude working until the condition you set is met
→ /rewind rolls back code and chat to a checkpoint (Esc twice does the same)
→ /code-review --fix finds the bugs in your diff and fixes them
→ /usage shows your cost and plan limits. /cost now opens it too
→ /plan makes Claude plan before it touches a single line
→ /batch splits a big change across parallel agents
→ /fork copies the chat into a background session while you keep going here
→ /compact summarizes the chat to free space, keeping what you name
→ /effort sets how hard Claude thinks, from low to max
→ /security-review scans your branch’s changes for vulnerabilities
→ /add-dir gives Claude one more folder for this session
Pick three, use them for a week, and the other ten will make sense on their own.
You don't need to pay to learn Claude
Seven Claude courses at GenAI Academy. All recorded, and all free.
Start with "What Can AI Actually Do For You?" if you're new, 30 minutes, no setup required.
Then the Claude Starter Course gets you running in 35.
Past that, pick by the problem you have. Hitting usage limits in Claude.ai, Cowork or Claude Code has its own course.
Building no-code agents has one.
So does rolling Claude out across an ops team in 30 days, from pilot to adoption data to full deployment.
The AI Portfolio Builder turns whatever you built into a case study you can show someone.
World Wide Vibes Hackathon is live too, $5,000 prize pool, beginner-ready, 100% online.
One login. Watch them in any order.
One developer just made Adobe optional
One developer recreated 7 of Adobe’s best-known apps with help from Claude Opus 5.5, all of them are free 😯
Open source and ready to download today.
Photoshop, Illustrator, Premiere, Lightroom, After Effects, InDesign and Acrobat Pro. Clean-room replacements, 100% native Rust.
No subscriptions. No cancellation fees. Fork them, build them, keep them.
The usual worry with a project like this is that it dies in six months.
The developer behind it has 15 years in the industry, writing six-nines payment rails moving billions a day, plus robotics and optics automation, and has used Rust for about a decade.
It isn’t a side project either. A team building software for filmmakers is funding the work, and there’s a test harness in place to prevent regressions as the apps move toward parity.
You can help get there. Clone the code, file issues, submit PRs, or flag gaps worth redesigning.
12 GitHub repositories every AI developer should know
1. OpenClaw (392K stars): Your own AI assistant that runs on your machine and answers you in WhatsApp, Slack or Telegram.
2. Hermes Agent (254K stars): Nous Research's agent that learns on the job. It turns finished tasks into skills and keeps improving them.
3. OpenCode (213K stars): Open-source coding agent for your terminal. Bring any model you want.
4. n8n (207K stars): Automate work across hundreds of apps with a visual editor and built-in AI.
5. Firecrawl (192K stars): Turn any website into a clean markdown or JSON your agents can actually use.
6. Ollama (183K stars): Run open models like Qwen, Gemma and DeepSeek on your own hardware.
7. Agent Skills (181K stars): Anthropic's library of Claude skills, including the ones Claude uses to make Word, Excel, PowerPoint and PDF files.
8. Dify (158K stars): Build agentic workflows and RAG apps on one platform, from prompt to deployment.
9. Open WebUI (156K stars): A ChatGPT-style interface you host yourself, for local or API models.
10. Langflow (156K stars): Drag and drop your way to a working agent, then ship it as an API.
11. Claude Code (151K stars): Anthropic's coding agent that works across your whole codebase from the terminal.
12. LangChain (148K stars): One of the most used frameworks for building agents. Skip the plumbing other teams have already solved.
Learning to use these well is what separates reading about AI from building with it.
GenAI Academy has free courses to get you started, including a practitioner's guide to AI agents with Claude.
Master Claude Code in 6 sessions
You already know the developers who seem to move twice as fast as everyone else.
They're not typing faster. They're:
Orchestrating teams of agents
Running parallel Claude instances across git worktrees
Wiring MCPs into their whole stack
Scheduling loops that keep working while they sleep
Almost none of it is written down in one place.
Mastering Claude Code course puts it all in one place.
6 live sessions, each one hands-on:
You start with the agentic loop and model selection
You build Skills and full agent teams from scratch
You apply Boris Cherny's 13 tips plus 6 new April 2026 techniques
You set up all 3 tiers of CLAUDE.md as a second brain that gets smarter every session
You wire MCPs into Slack, GitHub, Gmail and Supabase
You leave having built, tested and shipped your own custom Skill as a capstone
This is the deep track, the habits that separate casual users from power users.
Pricing is locked in for early sign-ups.
RAG, CAG and KAG solve the same problem in 3 very different ways
The problem: giving an LLM knowledge it was never trained on.
RAG (Retrieval-Augmented Generation)
→ Splits docs into chunks, stores them as vectors, and pulls the closest ones into the prompt at question time
→ Best for: large knowledge bases that change often
→ Watch out: if retrieval misses, the answer misses
CAG (Cache-Augmented Generation)
→ Loads all docs into the context window once, then reuses the saved KV cache for every question. No retrieval step
→ Best for: small, stable knowledge where speed matters
→ Watch out: capped by context size, and you re-cache whenever docs change
KAG (Knowledge-Augmented Generation)
→ Turns docs into a knowledge graph and answers by reasoning across it in steps
→ Best for: expert domains like law, medicine and finance, where every answer needs a trail
→ Watch out: graphs take real work to build and maintain
Quick rule:
→ Docs change daily? RAG
→ Small and stable? CAG
→ Questions that connect many facts? KAG
Many production systems mix them: a cached core for stable knowledge, retrieval for everything fresh.
Everything else you shouldn't miss
Krea 3 edits an image in 8 seconds: Six cents per image, four times faster than Nano Banana Pro, and it can blend up to eight references into one picture.
Nvidia’s new model thinks in fewer tokens: Nemotron 3.5 Super VL scored 26.5 on Artificial Analysis’ Intelligence Index while using substantially fewer output tokens than its predecessor. Nvidia also shipped the smaller Nemotron 3.5 Bolt 74B-A9B and near-lossless NVFP4 quantized weights.
GPT-Live-1.1 speaks six more languages: New voices in French, German, Hindi, Mexican Spanish, Korean and Japanese, with better accents and less of the constant checking in. Live now in ChatGPT, Codex and dots, with API access coming.
An 816B open model that runs on two Mac Studios: Meta’s Muse Spark Open has 42B active parameters and a 1M-token context window, built for coding and long-horizon agent tasks. Two Apple M5 Ultra Mac Studios run it locally through MLX.
Learn more about AI from the experts building it
Follow us on Instagram for fast, visual AI updates in 30 seconds.
Subscribe to our Atlas newsletter — trusted by 3M+ subscribers — to stay ahead of AI news across tech, education, and business.
📺 Watch us on YouTube to hear insights directly from leading AI voices, builders, and innovators.
🐦 Follow us on X for breaking AI news and real-time industry updates.
Learn how to build your next AI application with practical resources and expert guidance.
Explore investment opportunities in the future of AI and join our community-backed growth journey.
