Hey guys I just found this on X: https://github.com/cloudflare/security-audit-skill It’s a skill for doing a thorough security audit of your software I recommend installing it as a project level skill in your most important codebase and trying it out Of course, like any skill that you hear about online, you should first give to an agent and ask “is this skill safe to use?” This one is from cloudflare, so I’m sure it’s fine

Let me know what you're building with Jev! Please feel free to share: - the problem and why Jev solves it - a screenshot of the app/code - the schema of the request itself (e.g. response object with Choice / Score / Noul fields) (you don't have to share ALL of the above, those are just for inspiration)

Hi, To keep your local system clean, do you guys run agents in devcontainers or just run it locally? I know some people run their dev work in devcontainers where claude or codex is present so they can go crazy without accidentally creating junk on their system.

I made this so I can run commands in the terminal with less copy and pasting https://github.com/zazencodes/aicmd I'm looking for early users to test it out. Please help me out guys! 1. give the repo a star on GitHub ⭐ (thank youuuuuu) 2. install it and try it out curl -fsSL https://github.com/zazencodes/aicmd/raw/refs/heads/main/install.sh | sh I built this for myself and it works great on my system but I need your feedback It works with ollama, llama.cpp, APIs (openai, claude, gemini) and CLIs (codex, claude code, antigravity)

Read Gergely’s note (and excellent video) on the state of the industry (https://newsletter.pragmaticengineer.com/p/the-state-of-the-tech-industry-in), especially the bit where mid-size companies have built their own agent harnesses and cloud coding agents are what most devs will actually run. We are an AWS shop with a model gateway, LibreChat, and n8n already in place, and I am looking at Strands as a self-hosted harness in our own account. Two things I cannot tell from the outside: How are folks using cloud coding agents? We've mainly settled on OpenCode as a coding harness. Has anyone let internal users deploy their own agent harness, AWS Strands or otherwise, from a chat in a central platform account, without it becoming a second control plane? Curious what survived contact with security, not a benchmark.