On Programming with Agents

#1

Post by saurabh » Sat, Jan 24, 2026, 9:10 AM UTC

Re: On Programming with Agents

#2

Post by nezhar » Sat, Jan 24, 2026, 10:37 AM UTC

The first rule covers it best: it is crucial that you know what you are doing in order to benefit from these tools.

Regarding the second and third rules, I like to work with two terminals. One is for my agent, isolated in a container, and the other one I use for `git`. This way, I can assess whether my changes are moving in the right direction.

The PLAN.md files have also proven to be a good approach. I'm still trying to figure out the best way to keep them aligned with the application without polluting the context, but they hold significant value for documentation purposes.

Re: On Programming with Agents

#3

Post by saurabh » Sat, Jan 24, 2026, 2:37 PM UTC

nezhar wrote: post #2

The first rule covers it best: it is crucial that you know what you are doing in order to benefit from these tools. Regarding the second and third rules, I like to work with two terminals. One is for my agent, isolated in a container, and the other one I use for `git`. This way, I can assess whether my changes are moving in the right direction. The PLAN.md files have also proven to be a good approach. I'm still tryin…

I think a properly hand tuned AGENTS.md with proper tooling instructions and a single PLAN.md goes a long way especially if you add directives in AGENTS.md to keep the PLAN.md always updated with journaling.