How this skill is triggered — by the user, by Claude, or both
Slash command
/nebulaos:nebulaos-debugThe summary Claude sees in its skill listing — used to decide when to auto-load this skill
Use este guia para debugar execuções, analisar traces e investigar erros.
Use este guia para debugar execuções, analisar traces e investigar erros.
nebulaos exec list --status failed
nebulaos exec list --page-size 20
nebulaos exec get <execution-id>
nebulaos exec get <execution-id> -o json
nebulaos exec logs <execution-id>
nebulaos exec logs <execution-id> --verbose
nebulaos obs traces get --execution-id <execution-id>
nebulaos obs traces get --execution-id <execution-id> -o json
nebulaos obs traces search --agent meu_agent
nebulaos obs traces search --agent meu_agent --status error
nebulaos obs traces search --status error
nebulaos obs traces search --status success
nebulaos obs traces search --from 2026-03-01T00:00:00Z --to 2026-03-31T23:59:59Z
nebulaos obs traces search --limit 50
nebulaos obs traces search \
--agent meu_agent \
--status error \
--from 2026-03-01 \
--limit 20
nebulaos obs traces get <trace-id>
nebulaos obs traces get <trace-id> -o json
nebulaos obs metrics
nebulaos obs metrics --type latency
nebulaos obs metrics --from 2026-03-01 --to 2026-03-31
nebulaos exec logs <execution-id>
nebulaos llm usage
nebulaos llm requests --limit 50
nebulaos llm requests --route gpt4
# 1. Identificar
nebulaos exec list --status failed
# 2. Ver erro básico
nebulaos exec get <exec-id>
# 3. Ver logs
nebulaos exec logs <exec-id>
# 4. Trace completo
nebulaos obs traces get --execution-id <exec-id> -o json
# 5. Buscar padrões
nebulaos obs traces search --agent <agent-name> --status error
# IDs de execuções com erro
nebulaos exec list --status failed -o json | jq -r '.[].id'
# Último erro de um agent
exec_id=$(nebulaos exec list --status failed -o quiet | head -1)
nebulaos exec get $exec_id -o json | jq '.error'
npx claudepluginhub starya-nebulaos/nebulaos-claude-skills --plugin nebulaosQueries OpenSearch OpenTelemetry traces using PPL for GenAI agent invocations, tool executions, slow spans, errors, latency, and token usage via curl and Bash.
Fetches and analyzes LangSmith execution traces to debug LangChain/LangGraph agents. Use when investigating errors, tool calls, memory operations, or agent performance.