jwbron

20 ms·

1.
by jwbron 3mo ago
I'm using Claude code + (a patched) litellm proxy + openrouter + Qwen 3.7 max/kimi k2.6/deepseek v4 pro. The only feature that doesn't work is webfetch and web search, which I've replaced with the ddg MCP and a web
2.
by jwbron 3mo ago
I'm using Claude code + (a patched) litellm proxy + openrouter + Qwen 3.7 max/kimi k2.6/deepseek v4 pro. The only feature that doesn't work is webfetch and web search, which I've replaced with the ddg MCP. Memory, c
3.
Stop telling agents what not to do(github.com)
1 points by jwbron 6mo ago|0 comments
4.
by jwbron 7mo ago
The former GitHub CEO has a startup and this was their first release. They call it checkpoints: https://entire.io/ I copied it for my own tooling to make it work a bit better for my workflows.
5.
by jwbron 7mo ago
Important note: this thing eats tokens. I'm currently cycling through four different claude max 20x accounts, hitting weekly limits every couple of days.
6.
Egg: Intentional Agentic Developement(github.com)
1 points by jwbron 7mo ago|2 comments
7.
by jwbron 7mo ago
Egg is a side project of mine. It's a structurally enforced SDLC pipeline for autonomous LLM agents. You provide it with a task, it:refines the task definition and sends it back to you, you review the refinement, it generates a plan ba
8.
by jwbron 7mo ago
Right, but paying for food at a restaurant doesn't get me any closer to owning a restaurant. If promises are delivered on for these agentic coding platforms (which I do believe in), it seems the most reasonable path forward is to use t
9.
by jwbron 7mo ago
The software collects the network traffic of distributed Claude code instances.