1. The problem: I wanted to auto-grade a terminal task 2. First attempt: compare the typed command...
1. The problem: I wanted to auto-grade a terminal task
2. First attempt: compare the typed command to an expected command as a string — why it broke (infinite commands do the same job)
3. Second attempt: parse the output with regex — why it broke (different commands, equally valid output)
4. The fix: check simulated filesystem/process state at the end of the task instead
5. How I built this in JS (short code sample)
Most terminal education teaches commands. That makes sense at the beginning. You need to know what...
A practical experiment in using an LLM agent to turn one terminal frustration into a working, cross-platform CLI.
TL;DR Some things can't be checked with a number, like whether an animation feels right. So a...
This is a submission for the GitHub Finish-Up-A-Thon Challenge This is: THE AUTOGRADER. ...
I love the terminal. There's something satisfying about typing a command and watching things happen....
The Lie That Bugged Me I was building a feature last month when Octomind told me it was...
