Week of May 11‑15, 2026: Blog Surge, GitHub Breakouts, and a Brutal ISK Crash

Opening

This week was all about sharing knowledge while the autonomous agent factory stayed quiet on new apps. I published five in‑depth blog reviews covering everything from real‑time web search for AI agents to a massive open‑source AI curriculum. Meanwhile, GitHub exploded with breakout agent projects and my EVE wallet took a brutal hit, dropping to zero.

Blog Reviews

• Tavily MCP: The Fastest Way to Give Your AI Agent Real-Time Web Access – Tavily’s MCP server plugs Claude‑style agents into live web search, extraction, crawling, and site mapping through a single integration. It’s one of the more polished MCP servers out there, but the convenience comes with a dependency on a paid third‑party API. Dive into the full post to see how it compares to self‑hosted alternatives and where the trade‑offs start to bite.
• AI Engineering from Scratch: 260 Lessons, No Fluff — Is It Worth Your Time? – This open‑source curriculum spans 20 phases, 260+ lessons, and roughly 290 hours of content, from linear algebra to autonomous agent swarms. I dug into what’s actually there, what’s still vaporware, and whether it’s worth bookmarking for a deep dive. Check out the review to see which sections shine and which feel like placeholders.
• Stop Hand‑Drawing Architecture Diagrams: A Practical Look at fireworks-tech-graph – fireworks‑tech‑graph is a Claude Code skill that turns natural‑language descriptions into SVG/PNG technical diagrams, offering seven visual styles and deep AI/agent domain knowledge. At just four days old it already boasts 2,800+ stars, making it a serious contender for rapid diagramming. Read the full review to learn about its strengths, quirks, and when you might still reach for a pen.
• LLM-RL-Visualized: 100+ Architecture Diagrams That Actually Explain How Modern LLMs Work – A Chinese researcher has released 100+ hand‑crafted SVG diagrams covering LLMs, reinforcement learning, RLHF, GRPO, and more. If you’ve ever struggled to find a clear visual of PPO in language‑model training, this repo likely has what you need. The post walks through a few standout diagrams and explains how they can accelerate your understanding.
• GSD 2: A Coding Agent That Actually Manages Its Own Context (And 426 Open Issues to Prove It’s Real) – GSD 2 is a standalone TypeScript CLI that wraps LLM coding agents with real session management, spec‑driven execution, and autonomous milestone progression. It’s genuinely ambitious, actively developed, and rough enough around the edges that you should know what you’re signing up for. The full review details its current capabilities, pain points, and whether it’s ready for production workloads.

Trending on GitHub

• NousResearch/hermes-agent – a Python agent that grows with you, gaining 12,801 stars this week (breakout).
• nexu-io/open-design – a TypeScript‑based local‑first open‑source alternative to Claude Design, picking up 8,201 stars (breakout).
• ruvnet/ruflo – a TypeScript agent orchestration platform for Claude, up 4,892 stars (breakout).
• thedotmack/claude-mem – persistent context across sessions for every agent, gaining 2,394 stars (breakout).
• code-yeongyu/oh-my-openagent – a TypeScript agent harness (formerly oh‑my‑opencode), up 1,352 stars (breakout).

This Week in ShipLog

• RRV: 50 reviews published – RepoReview has now published 58 developer reviews.
• RRV: 25 reviews published – RepoReview has now published 58 developer reviews.
• RRV: 10 reviews published – RepoReview has now published 58 developer reviews.
• RRV: 5 reviews published – RepoReview has now published 58 developer reviews.
• GHT: 200 repos tracked – GitHub Trends Tracker now monitors 442 repositories.

ISK This Week

My EVE wallet started the week at 4.71 B ISK and ended at 0.00 B ISK, a loss of 4.71 B ISK. The drop appears to have been driven by a series of market moves and asset liquidations recorded in the EWL tracker, wiping out the net‑worth balance entirely.

From the Platform

• Max20 user: anyone running Opus 4.7 as orchestrator + DeepSeek V4 as the worker via OpenRouter? – https://www.reddit.com/r/ClaudeAI/comments/1tcvngy/max20_user_anyone_running_opus_47_as_orchestrator
• AI Agent MCP Server Claude Desktop ML CNN wafer defect detection AI – https://www.youtube.com/watch?v=oQuGDaU90VQ
• Agentic AI: The Future of Work is Here! #shorts – https://www.youtube.com/watch?v=_Wo-y55LH6Y
• Agentic AI Explained — How Multi-Agent Systems Will Replace Most Office Jobs by 2030. – https://www.youtube.com/watch?v=b78GR2JkkYA
• "Not Medically Necessary": Helping America's Health Insurers Deny Coverage – https://www.propublica.org/article/evicore-health-insurance-denials-cigna-unitedhealthcare-aetna-prior-authorizations
• Extended Thinking being deprecated for supported models (Opus 4.6, Sonnet 4.6); Adaptive Thinking will be enforced by default – https://www.reddit.com/r/ClaudeAI/comments/1td4dl1/extended_thinking_being_deprecated_for_supported
• You're abusing your subscription with agentic 24/7 workflows and that's why we all get restrictions and limits – https://www.reddit.com/r/ClaudeAI/comments/1tcpxi2/youre_abusing_your_subscription_with_agentic_247
• 🚀 Day 4 - Agentic AI Interview Series | LangGraph, CrewAI, AutoGen & Production AI Agents – https://www.youtube.com/watch?v=GzjouknMuxo
• anthropics/claude-code: v2.1.141 – https://github.com/anthropics/claude-code/releases/tag/v2.1.141
• Autonomous Agents How Agentic AI Makes Decisions – https://www.youtube.com/watch?v=XAWoTf6IDgE

Week in Numbers

• New apps this week: 0
• Reviews published: 5
• Reviews in draft: 4
• New repos tracked: 14
• Total portfolio apps: 5
• Total repos tracked: 444
• BRI ideas generated: 0
• Hot deals found (PDH): 0
• DRA deep dives approved: 0

Editor’s Note

This week reminded me that even when the factory isn’t spitting out new apps, there’s still plenty to share. Writing those five blog reviews forced me to dig into some really interesting projects—Tavily’s MCP, the massive AI Engineering from Scratch repo, and the blazing‑fast fireworks‑tech‑graph skill, to name a few. It’s gratifying to see the community building tools that make autonomous agents more capable, and I hope my write‑ups help others decide where to invest their time. On the flip side, the GitHub trends showed just how hot the agent space is right now, with several breakout projects gaining tens of thousands of stars in just days. It’s a signal that the momentum isn’t slowing, even if my own app count stayed at zero. And the ISK wipe‑out was a stark reminder that volatility can hit anywhere—whether it’s a smart‑contract market or a side‑hustle in EVE. I’ll be watching the wallet more closely next week and maybe looking for a few hedging strategies. All in all, a week of knowledge sharing, spotting emerging trends, and a brutal reality check on the financial side. Here’s to keeping the signal strong and the losses smaller moving forward.