Adds _find_session_uuid() to look up a session's UUID from Claude's JSONL files
_run_loop() now shows claude --resume <uuid> (the actual UUID) when an agent is still running, rather than the session name which doesn't work with --resume
Falls back to pointing at scripts/agent_loop.py list when the UUID isn't resolved yet
_start_agent() no longer prints a broken resume command at startup (UUID not available immediately); directs user to list subcommand instead
Module docstring updated to document the correct UUID-based resume workflow
13 new tests
Test plan
All 50 Python unit tests pass (python3 -m unittest scripts/test_agent_loop.py)
Pre-commit hooks pass (dart format, large-file check, etc.)
## Summary
- Adds `_find_session_uuid()` to look up a session's UUID from Claude's JSONL files
- `_run_loop()` now shows `claude --resume <uuid>` (the actual UUID) when an agent is still running, rather than the session name which doesn't work with `--resume`
- Falls back to pointing at `scripts/agent_loop.py list` when the UUID isn't resolved yet
- `_start_agent()` no longer prints a broken resume command at startup (UUID not available immediately); directs user to `list` subcommand instead
- Module docstring updated to document the correct UUID-based resume workflow
- 13 new tests
## Test plan
- [x] All 50 Python unit tests pass (`python3 -m unittest scripts/test_agent_loop.py`)
- [x] Pre-commit hooks pass (dart format, large-file check, etc.)
- [ ] CI on the self-hosted runner
Fixes #152
Blocking a user prevents them from interacting with repositories, such as opening or commenting on pull requests or issues. Learn more about blocking a user.
Summary
_find_session_uuid()to look up a session's UUID from Claude's JSONL files_run_loop()now showsclaude --resume <uuid>(the actual UUID) when an agent is still running, rather than the session name which doesn't work with--resumescripts/agent_loop.py listwhen the UUID isn't resolved yet_start_agent()no longer prints a broken resume command at startup (UUID not available immediately); directs user tolistsubcommand insteadTest plan
python3 -m unittest scripts/test_agent_loop.py)Fixes #152