ai-management-maintenance

Enforce database-first priority-based AI provider selection with Ollama fallback.

1|1|Updated Jul 20, 2025

Search Tags:#monitoring #analytics #orchestration #fallback #model-discovery #ai-providers #ai-management

Ready to run

npx skills add https://github.com/mdresch/adpa --skill ai-management-maintenance

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: ai-management-maintenance
Source: https://github.com/mdresch/adpa/tree/main/ai-management-maintenance
Command: npx skills add https://github.com/mdresch/adpa --skill ai-management-maintenance

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

Maintains and troubleshoots the ADPA AI Management system to ensure reliable provider selection, robust fallbacks, and consistent analytics across deployments.

Core Features & Use Cases

Ensures a database-first, priority-based AI provider selection to honor custom configurations.
Enforces a mandatory Ollama fallback to local models for offline resilience.
Supports model discovery and synchronization workflows, including updating available_models and default_model, with backward-compatible configuration in JSONB.
Provides real-time analytics checks and maintenance procedures to keep dashboards accurate and responsive.

Quick Start

Configure a new AI provider in the admin console, verify that fallbacks trigger on simulated outages, and validate analytics data integrity.

Frequently Asked Questions about ai-management-maintenance

High-intent search queries and answers about installing and using this skill.

FAQPage Schema

How do I configure a database-first priority policy for AI provider selection?▼

Database-first AI provider selection enforces custom priority configurations by resolving models from a central database before applying fallbacks. This ensures custom configurations are honored consistently across deployments.

Why does my AI provider selection fail during network outages?▼

AI provider selection fails without offline resilience if no local fallback is configured. Enforcing a mandatory Ollama fallback to local models maintains service availability when external providers experience outages.

How does model discovery synchronization update available models?▼

Model discovery synchronization updates available_models and default_model entries using backward-compatible JSONB configuration. This maintains consistent model availability across your AI management system.

What is the best way to validate real-time AI usage analytics integrity?▼

Validating real-time AI usage analytics requires running maintenance procedures and integrity checks to verify dashboard data. This keeps analytics dashboards accurate and responsive across deployments.

Can I use JSONB configuration for backward-compatible AI model updates?▼

Yes, JSONB configuration supports backward-compatible model discovery and synchronization workflows. This allows updating available_models and default_model without breaking existing AI management configurations.