OperationsTroubleshooting
Performance Issues
Reduce response latency and long task execution time
Performance Issues
Typical bottlenecks
- Slow model responses
- Overlapping scheduled tasks
- Large command outputs
- Long conversation context
Optimization checklist
- Use faster models for high-frequency interactions.
- Reduce task frequency to avoid overlaps.
- Cap command output size:
{
"permissions": {
"exec_command": {
"maxOutputChars": 12000,
"maxOutputLines": 200
}
}
}- Tune context compaction:
{
"context": {
"messages": {
"maxInputTokensApprox": 128000,
"compactRatio": 0.5
}
}
}Observe performance signals
grep -Ei "timeout|slow|context|compact" .ship/logs/$(date +%F).jsonl