TOOLS

–

11 April 2026 19h05

If there’s an open format that harnesses use (which right now there’s not)

→ View original post on X — @garymarcus

Tools and Harnessing Drive Recent AI Progress Over Scale

By

@garymarcus

–

11 April 2026 18h56

nope, tools/harnessing is what is accounting for most of the recent progress. not scale. no way could you get a pure LLM to do what Claude Code does. that’s why they used the other stuff, too see eg https://
open.substack.com/pub/garymarcus
/p/how-o3-and-grok-4-accidentally-vindicated?r=8tdk6&utm_medium=ios
…

API Provider Features and Prompt Caching in Deep Agents

By

–

11 April 2026 18h54

this is a good point around taking advantage of model/api provider features i agree that prompt caching is great! we make sure to use it in deepagents! but that alone doesnt lock in – you can switch pretty easily, just a bit more costly things like encrypted or serverside

Understanding Agent Harnesses in LangChain Development

By

–

11 April 2026 18h52

@sarahwooders helped educate me on some of the lock in as well https://
blog.langchain.com/the-anatomy-of
-an-agent-harness/
… is a great blog by @Vtrivedy10 on agent harnesses i think the deep agent docs are pretty good as well

→ View original post on X — @cursor_ai

Composer 2 Usage Doubled in New Interface Update

By

@cursor_ai

–

11 April 2026 18h52

This weekend we're doubling Composer 2 usage in our new interface.

Open Agents Window, pick Composer 2, and start building! No hourly limits.https://t.co/Z5GHz7kcG1
— Cursor (@cursor_ai) 11 avril 2026

This weekend we're doubling Composer 2 usage in our new interface. Open Agents Window, pick Composer 2, and start building! No hourly limits.

Prompt Caching vs Vendor Lock-in in AI Agents

By

–

11 April 2026 18h50

i agree that prompt caching is great! we make sure to use it in deepagents! but that alone doesnt lock in – you can switch pretty easily, just a bit more costly things like encrypted or serverside compaction make it harder to do so. blackbox long term memory makes it impossible

Thoth: Advanced Agent Harness with State-of-the-Art Memory

By

–

11 April 2026 18h45

check out thoth – agent harness with sota memory built on langgraph

→ View original post on X — @levelsio

Using Hetzner VPS with Tailscale, Termius, Claude Code, and Mosh

By

@levelsio

–

11 April 2026 18h44

I use @Hetzner_Online and just have VPS that are isolated via Tailscale (for safety) then I SSH into them with Termius and work on there with Claude Code and I often have Mosh running so it stays alive easier