[Bug] Subagents using Python comments as internal communication mechanism

Bug Description
I've noticed that during code review cycles or when subagents are running conclusions that they will lean towards emitting a `python3 -c "<multiline commented out prose>" kind of thing to basically communicate with itself, which seems very wrong to me.

Environment Info

• Platform: darwin
• Terminal: iTerm.app
• Version: 2.1.87
• Feedback ID: 7f810320-63b0-43e3-83ba-3dd8b36a44ef

Errors

[{"error":"AxiosError: Request failed with status code 401\n    at _B (/$bunfs/root/src/entrypoints/cli.js:144:1194)\n    at <anonymous> (/$bunfs/root/src/entrypoints/cli.js:149:12745)\n    at emit (node:events:92:22)\n    at endReadableNT (internal:streams/readable:861:50)\n    at processTicksAndRejections (native:7:39)\n    at request (/$bunfs/root/src/entrypoints/cli.js:151:2467)\n    at processTicksAndRejections (native:7:39)","timestamp":"2026-04-01T12:54:56.021Z"},{"error":"AxiosError: Request failed with status code 401\n    at _B (/$bunfs/root/src/entrypoints/cli.js:144:1194)\n    at <anonymous> (/$bunfs/root/src/entrypoints/cli.js:149:12745)\n    at emit (node:events:92:22)\n    at endReadableNT (internal:streams/readable:861:50)\n    at processTicksAndRejections (native:7:39)\n    at request (/$bunfs/root/src/entrypoints/cli.js:151:2467)\n    at processTicksAndRejections (native:7:39)","timestamp":"2026-04-01T12:54:56.097Z"}]