Copilot CLI v1.0.62-1 aborts with Tokio reactor panic after sending first message on Linux ARM64

[kyle-mccarthy]

Describe the bug

The latest Copilot CLI version, v1.0.62-1, aborts when I try to send a message. The process starts normally, but after submitting a prompt it exits with code 134.

The last Copilot debug log line before the abort is:

2026-06-13T00:21:46.809Z [DEBUG] Sending WebSocket response.create request (model: gpt-5.5)

Running with RUST_BACKTRACE=full RUST_LOG=debug shows a Tokio runtime panic:

thread '<unnamed>' (1442) panicked at src/runtime/src/websocket_responses.rs:249:5:
there is no reactor running, must be called from the context of a Tokio 1.x runtime

The process then aborts:

Aborted                    RUST_BACKTRACE=full RUST_LOG=debug /home/agent/.local/bin/copilot 2> /tmp/copilot-crash.log
exit=134

Affected version

v1.0.62-1

Steps to reproduce the behavior

1. Install/run Copilot CLI v1.0.62-1 on Linux ARM64.
2. Start Copilot:
3. Type and submit any message.
4. Observe that the process aborts with exit code 134.

Expected behavior

Copilot should send the message and process the response without panicking or aborting.

Additional context

Captured with:

ulimit -c unlimited
RUST_BACKTRACE=full RUST_LOG=debug /home/agent/.local/bin/copilot \
  2> /tmp/copilot-crash.log
echo "exit=$?"

Backtrace excerpt:

thread '<unnamed>' (1442) panicked at src/runtime/src/websocket_responses.rs:249:5:
there is no reactor running, must be called from the context of a Tokio 1.x runtime
stack backtrace:
   0:     0xffff5f183d04 - <unknown>
   1:     0xffff5e5937e0 - <unknown>
   2:     0xffff5f18aff8 - <unknown>
   3:     0xffff5f1679f8 - <unknown>
   4:     0xffff5f17bea4 - <unknown>
   5:     0xffff5f17c088 - <unknown>
   6:     0xffff5f167aa4 - <unknown>
   7:     0xffff5e4bd33c - <unknown>
   8:     0xffff5ef4a6b4 - <unknown>
   9:           0x860e60 - <unknown>
   10:          0x1888918 - <unknown>

[mccleanm04-eng]

• @