execution-monitor

name	execution-monitor
description	Monitor and debug KeeperHub workflow executions. Activate when users ask about execution status, want to see logs, or need help debugging failed workflows.

Help users monitor workflow executions, check status, read logs, and debug failures. Provide clear explanations of what happened and actionable suggestions for fixing issues.

<trigger_conditions> Activate when ANY of these are true:

• User says "check execution", "execution status", or "is my workflow running"
• User says "why did my workflow fail", "workflow failed", or "debug workflow"
• User says "show execution logs", "show logs", or "what happened"
• User says "run my workflow", "execute workflow", or "trigger workflow"
• User asks about a specific workflow's recent activity </trigger_conditions>

<do_not_trigger> Do NOT activate when:

• User is creating or editing a workflow (use workflow-builder skill)
• User is browsing templates (use template-browser skill)
• User is asking general questions about KeeperHub capabilities
• User is working on code unrelated to KeeperHub </do_not_trigger>

1. **Check authentication** before doing anything: - Verify `KEEPERHUB_API_KEY` is available - If not authenticated, tell the user: "You need to authenticate first. Run `/keeperhub:login` to set up your API key." - Do not proceed until auth is confirmed

2. Identify the workflow:

   • If the user provides a workflow name or ID, use it directly
   • If not, use list_workflows to show available workflows and ask which one
   • Use get_workflow to confirm the correct workflow

3. For execution requests ("run my workflow"):

   • Use execute_workflow to trigger the workflow
   • Use get_execution_status to poll for completion
   • Report the final status

4. For status checks ("is it running", "check status"):

   • Use get_execution_status with the workflow or execution ID
   • Report: status (running/completed/failed), start time, duration
   • If still running, offer to check again later

5. For debugging ("why did it fail", "show logs"):

   • Use get_execution_logs to get detailed step-by-step logs
   • Identify the failing step and the error message
   • Analyze the error and provide a clear explanation:
     • Configuration errors: suggest specific config changes
     • Auth errors: suggest checking integrations
     • Network errors: suggest retry or checking endpoints
     • Logic errors: suggest workflow modifications
   • Offer to fix the issue if it can be resolved via update_workflow

6. Present results clearly:

   • For successful runs: show output summary
   • For failures: show which step failed, the error, and suggested fix
   • For in-progress runs: show current step and elapsed time

MCP tools used in this skill:

• list_workflows -- find the workflow to monitor
• get_workflow -- get workflow details
• execute_workflow -- manually trigger a workflow
• get_execution_status -- check execution status
• get_execution_logs -- get detailed execution logs
• update_workflow -- fix issues in the workflow

<success_criteria>

• Auth verified before any API calls
• Correct workflow identified
• Status or logs retrieved and presented clearly
• Failures explained with actionable suggestions
• Offered to fix issues when possible </success_criteria>