In partnership with
Hey {{first_name | there}},
Claude has really upped the game lately.
Between Claude Skills, better coding abilities, and now a growing ecosystem of plugins, it’s no longer just an AI assistant.
It’s starting to feel like a full dev environment.
And if you’re not using plugins inside Claude Code, you’re honestly missing out.
I sat down and filtered through a bunch of them. Here are the ones that are actually useful:
Best Claude Code Plugins Worth Trying
Code Review — You can automate your entire PR review process. It runs 5 AI agents that catch bugs and filter false positives before you even look at the code. Type /code-review
Frontend Design — It will help you build interfaces that actually look designed, not AI-generated. Bold choices, real animations, unique components. Auto-triggers on any frontend work, no command needed.
Context7 — You can pull version-specific docs straight from source repos. Works with React, Prisma, and hundreds more. Type /context7
Playwright — Use this when you want to test your app by just describing what to click in plain English. No Selenium, no scripts. Type /playwright
Superpowers — It will help you make Claude actually plan before it touches your code. Brainstorm, plan, then execute with TDD built in. Type /superpowers
Backend Development — Use this when you need production-ready patterns for GraphQL, workflow orchestration, and test-driven backend dev without starting from scratch. Type /backend-dev
How to install:
Official plugins: /plugin install [name] directly in Claude Code
Community plugins: npx claude-plugins install [name] in your terminal
🛠️ AI Tools Worth Checking Out
Excelmatic: Analyze and visualize Excel data using plain English, no formulas needed.
AdMake AI: Spy on competitor ads and generate high-converting creatives in minutes.
Genspark: Hire an AI “employee” that runs tasks across apps, manages workflows, and gets work done for you.
Tosea AI: Turn academic papers or PDFs into clean, presentation-ready slides instantly.
AI News You Should Know
Anthropic used a version of Claude, the Claude interviewer, to run 81,000 interviews across 159 countries in 70 languages, all within a week.
The biggest thing people want from AI is simple. They want it to handle boring work so they can focus on better work and life. Time freedom, financial independence, and managing daily life came up a lot.
On the other side, the biggest concern was not job loss. It was AI getting things wrong, followed by losing control and becoming too dependent on it.
Why it matters: Standard polls give you a number. This study gives you texture. But the bigger story might be the method itself. Claude ran 81,000 in-depth interviews across 70 languages in one week. That kind of research scale simply did not exist a year ago. It is a strong proof of concept for AI as a serious research tool beyond writing code or summarizing text.
Someone found 3 instructions in Anthropic’s docs that drastically improve accuracy:
Tell Claude it is okay to say “I don’t know”
Require citations for claims
Make it use direct quotes from sources
Using all three together makes outputs much more reliable.
There is a tradeoff. It reduces creativity. The smarter way is to treat this as a research mode and turn it on only when needed.
Why it matters: These are not hacks. They are in Anthropic's official docs and most people have never read them. If you are doing research, fact-checking, or any kind of document analysis with Claude, these three instructions belong in your system prompt.
OpenAI plans to grow from around 4,500 to 8,000 employees by 2026, hiring across engineering, research, product, and sales.
They are also adding roles focused on helping companies actually use AI, not just buy access to it.
This comes as more businesses are choosing Anthropic over OpenAI for first-time AI adoption.
Why it matters: Anthropic's growth has gotten loud enough that OpenAI is responding with a serious headcount push. The AI race is not cooling down. Both platforms are going to be significantly more resourced by end of year, which means tooling, features, and competitive pressure all keep accelerating.
Microsoft released MAI-Image-2, now ranked among the top image models globally.
It focuses on what actually matters in real use cases. Better realism, better text inside images, and stronger control over complex scenes.
It is already rolling out across Copilot and Bing.
Why it matters: Microsoft now has a top-three image model and distribution across Copilot, Bing, and Azure. For developers building anything that needs image generation, MAI-Image-2 is now a serious option to test. The in-image text generation is a real differentiator for anyone building designed outputs like slides, posters, or branded content.
Nvidia CEO Jensen Huang proposed giving engineers AI tokens, or compute credits, on top of their salary.
The idea is simple. More tokens means more AI agents, which means more output per engineer.
He also described a future where companies could have more AI agents than human employees.
Why it matters: Token budgets as part of compensation is a genuinely new idea, and Huang does not float things casually. The engineer who knows how to deploy and manage AI agents effectively is going to be worth a lot more than one who does not. Companies are starting to price that difference in, and this week made it more visible than it has been before.
That is a wrap for this week.
Plugins are still one of the most underused parts of Claude Code. Most people are using it like a chatbot. The ones using these tools are treating it like a full team. That gap is going to get wider fast.
The news this week points in one direction: things are moving faster than most people expect, and the people who figure out how to actually use these tools instead of just knowing they exist are the ones who are going to feel it first.
If you try any of these plugins, reply and let me know what actually moved the needle for you. Always curious what works in the real world versus what just looks good on paper.
See you next week.
Stay building.
- Aashish
The Future of AI in Marketing. Your Shortcut to Smarter, Faster Marketing.
Unlock a focused set of AI strategies built to streamline your work and maximize impact. This guide delivers the practical tactics and tools marketers need to start seeing results right away:
7 high-impact AI strategies to accelerate your marketing performance
Practical use cases for content creation, lead gen, and personalization
Expert insights into how top marketers are using AI today
A framework to evaluate and implement AI tools efficiently
Stay ahead of the curve with these top strategies AI helped develop for marketers, built for real-world results.