Goodbye Claude, Hello Pi

Re:Goodbye Claude, Hello Pi by siraben on Sunday November 08, 2026 @12:41AM (#50410725)

I've been daily driving Pi for over a month now and never looking back. The latest crop of frontier models have really complex bash calls, so I've written an extension that rewrites them into plain English while still making them easy to reveal in full, unlike some harnesses which seem to try to hide this from the user which is pretty irresponsible IMO.[0]

[0] https://github.com/siraben/pi-tool-summaries

[ Parent | on HN ]

Re:Goodbye Claude, Hello Pi by lifeisloving on Sunday November 08, 2026 @12:53AM (#50410770)

I just have a hook that tells agents to echo what they're trying to do. Works fine.

[ Parent | on HN ]

Re:Goodbye Claude, Hello Pi by someothherguyy on Sunday November 08, 2026 @01:54AM (#50411024)

you could just learn to read it

[ Parent | on HN ]

Re:Goodbye Claude, Hello Pi by siraben on Sunday November 08, 2026 @09:19AM (#50413592)

The rewrite is configurably off for bash/tool calls below a certain line length. I would prefer not to read codegolfed bash and python for hours every day, personally.

[ Parent | on HN ]

Re:Goodbye Claude, Hello Pi by alex_duf on Sunday November 08, 2026 @04:24AM (#50411658)

I just have custom instructions to tell the model to edit files using its tools rather than a gigantic bash pipelines. That's 80% of it.

[ Parent | on HN ]

Re:Goodbye Claude, Hello Pi by machomaster on Sunday November 08, 2026 @05:01AM (#50411850)

Would that be useful in other harnesses/models like Codex? Wouldn't that lead to increased token usage?

[ Parent | on HN ]

Re:Goodbye Claude, Hello Pi by siraben on Sunday November 08, 2026 @09:19AM (#50413588)

I prefer not the fight the RL as the models get better, unless I've measured it and haven't observed any degradation. There may be other costs/no. of turns considerations too.

[ Parent | on HN ]