1. Show HN: Stop Using LangChain (synapsekit.github.io)
Async-first Python framework for building production-grade LLM apps. RAG, agents, graph workflows. pip install synapsekit.
2. Show HN: Leaking AWS Keys from a Trading Competition Sandbox via stdout (github.com)
Technical analysis and Proof of Concept (PoC) regarding environment variable exfiltration in containerized cloud sandboxes via side-channel data leaks. - JakOb-dotcom/cloud-sandbox-security-analysis
3. Show HN: The RUM Conjecture Explained [video] (www.youtube.com)
The RUM Conjecture proves that optimizing reads, writes, and memory is a three-way tradeoff - push one down, another goes up. Every engine on the market is m...
4. Show HN: Arrow. Point your phone, walk where it says, find out where (kouh.me)
Point your phone. Walk where it says. It
5. Show HN: EDDI – Multi-agent AI engine where agent logic lives in JSON, not code (github.com)
Config-driven engine that turns JSON into production-grade AI agents. Multi-agent orchestration, 12+ LLM providers, MCP/A2A protocols, RAG, persistent memory, and enterprise compliance (EU AI Act, ...
6. Show HN: I'm building Eigen, a self-hosted Google Workspace alternative (reindernijhoff.net)
Six months ago I wrote about Eigen: a European Google Workspace alternative where you own your data. Back then I had five basic apps, built in two months of spare time. Mail, drive, docs, contacts and Trello-like kanban boards. Minimal, but functional.
8. Show HN: SafeWeave – 8 security scanners as an MCP server inside your AI editor (safeweave.dev)
8 scanners. One MCP command. Finds vulnerabilities in 12 seconds. Built for Cursor, Claude Code & VS Code.