GPT 5.6 luna times out when re-steered

[MythicSorcerer]

Summary

Command Code never gets a response from the model when I hit esc to stop the model and give it a second instruction.
Actually version: 1.38.1, MacOS 26.0 but I ssh-ed into that machine and popups don't work across ssh so the stuff below is wrong.

Expected Behavior

The model should respond

Actual Behavior

The model doesn't respond

Steps to reproduce the issue

1. /model gpt 5.6 luna
2. Tell it do do something
3. Immediately or after a while hit esc to stop it
4. Ask it to do something else too
5. Get unlucky

Command Code Version

1.36.0

Operating System

macOS

Terminal/IDE

ghostty

Shell

zsh

Session file (optional)

{"type":"session","version":3,"id":"5c5c2695-13e9-4387-9ab7-7d4ee2d223c8","timestamp":"2026-08-28T20:07:38.636Z","cwd":"/Users/"}
{"type":"message","id":"0c983bd1","parentId":null,"timestamp":"2026-08-28T20:10:21.020Z","message":{"role":"user","content":[{"type":"text","text":"ollama launch pi pretends to have a 128k context window but Response was truncated before completion. ↑7.1k ↓3.0k 2.5%/128k (auto) ... so... can we fix that? can we make the output limit higher?"}],"meta":{"source":"user","createdAt":1787947765652,"messageId":"49e79408-6bd5-4c1a-90fe-dfac55c77639"}}}
{"type":"message","id":"74a6e39c","parentId":"0c983bd1","timestamp":"2026-08-28T20:10:21.021Z","message":{"role":"user","content":[{"type":"text","text":"or should we switch to llama-cpp"}],"meta":{"source":"user","createdAt":1787947785651,"messageId":"73e30cec-abad-4f24-ad89-0707ab3a7ebc"}}}
{"type":"message","id":"2a35d34a","parentId":"74a6e39c","timestamp":"2026-08-28T20:10:21.022Z","message":{"role":"user","content":[{"type":"text","text":"Error: Connection error.\n\nType "continue" to try again. If the issue persists, contact support: https://commandcode.ai/discord\nTrace ID: 01e4454d312cf7e77bf1685902168e91"}],"meta":{"messageId":"22c6552b-6d38-4609-b02b-fe558708cc45"}}}

Fix prompt (optional)

No response

Additional context

OS: macOS 27.0