Here's a quick tool for anyone who wants a way to simplify error messages with AI without needing to switch to a fully agentic terminal solution. Simply add hank before compiling/running your program to get a clear, readable explanation of any errors that occur.
You can think of this as just another system utility like grep, ls or echo that is at your fingertips for debugging a program. It uses local models as well so your code outputs don't go into online LLM services.
Excited to see what gets thrown back, hope it helps someone. Feel free to add onto it as its pretty raw at the moment.
petargyurov•3mo ago
Regarding your README, I would add a text version of the example in the demo.
alsetmusic•3mo ago
At the very least, it’s a wildcard. I think you’re not wrong about setting an alias (I have an alias for `??` to link as `$?`, for example). I just think the single character usage will create conflicts and unexpected outcomes.
Cool tool, OP. I’ve marked it to check out when I’m at my computer at home.
dillondesilva•3mo ago
As @alsetmusic already mentioned, I think an alias might be the best way for the moment.
One thing that could be challenging is figuring out how to access the stdout/stderr contents for a program that was previously run. With the current implementation, these pipes are routed to hank although I'm not immediately too sure how one could access this post-execution. Feel free to write or implement any ideas onto the repo!
alsetmusic•3mo ago
What luck that I still had this on my to-do list after seeing a couple of posts that coincide with this notion the other day:
https://news.ycombinator.com/item?id=45682219 https://news.ycombinator.com/item?id=45679839
TLDR: capture output for reuse / run commands on output without regenerating it.