Give your agent a sandbox
An LLM that writes code is only useful if it can run the code. Handing it your shell is one answer,...
#agents
4 posts tagged with agents
Agents are jobs, not services
A resident agent idling at 2GB of RAM is nothing, expensively. Most of the agent work people...
Lessons from building an agent harness for local models
The loop, the context juggling, the per-model prompt formatting, the tool calls. Here is what...
The Missing Playbook for AI Agents
Most people are building agents the way they wrote their first prompts. But we already know how to...
March 17, 2026
·
20 min read