Diagnoses and resolves issues with Olakai AI agent monitoring, KPI calculations, and SDK integrations.
The olakai-troubleshoot skill provides a robust diagnostic framework for developers using the Olakai AI monitoring platform. It implements a specialized 'Test → Fetch → Validate' workflow to systematically identify the root cause of observability gaps, such as missing events, malformed KPI data, or authentication failures. By providing contextual CLI commands and code-level fixes, this skill helps users resolve common errors involving KPI formula parsing, case-sensitive custom data mismatches, and SDK initialization hurdles to ensure accurate agent performance tracking.
Key Features
01KPI formula validation and AST parsing corrections
02Authentication and API key integrity verification
03Case-sensitivity checks for customData and schema mappings
04Direct API health and connectivity testing
050 GitHub stars
06Automated diagnostic workflows for SDK and CLI issues
Use Cases
01Resolving agent-specific KPI availability and configuration errors
02Fixing KPI values that display as strings or nulls instead of numbers
03Debugging why events are not appearing in the Olakai dashboard
