rag-service-maintenance

Re-index ChromaDB vault, skills, and sessions, then restart the RAG service with health checks.

Updated Apr 19, 2026
One-click install
npx skills add https://github.com/crycriM/hermes-skills --skill rag-service-maintenance
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: rag-service-maintenance
Source: https://github.com/crycriM/hermes-skills/tree/main/mlops/rag-service-maintenance
Command: npx skills add https://github.com/crycriM/hermes-skills --skill rag-service-maintenance

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Re-indexes vault, skills, and sessions into ChromaDB and restarts the RAG service to keep results accurate and the service healthy after content updates.

Core Features & Use Cases

  • Re-index memory content (vault, skills, sessions) into ChromaDB to refresh knowledge base.
  • Backfill new conversations and ensure search indices reflect latest data.
  • Restart and validate the RAG service to maintain availability in production.

Quick Start

Run the maintenance cycle to re-index vault content, refresh sessions, and restart the RAG service with health checks.

Frequently Asked Questions about rag-service-maintenance

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

FAQPage Schema
How do I re-index ChromaDB after adding new content to my RAG service?

To re-index ChromaDB after content updates, run the Vault Indexer and Session Backfill to refresh vault, skills, and sessions data, ensuring your RAG service returns up-to-date search results.

What is the best way to restart a RAG service and verify its health?

The best way to restart a RAG service involves using systemctl for a controlled restart followed by health verification via curl and log inspection to maintain availability in production.

Why does my RAG service return outdated search results after a content update?

Outdated RAG search results occur because the ChromaDB indices are stale; re-indexing vault, skills, and sessions backfills the new conversations and refreshes the knowledge base.

When do I need to run a maintenance cycle on my RAG service?

You need to run a RAG maintenance cycle after content updates or when new material is added, ensuring the ChromaDB vault and sessions are re-indexed to reflect the latest data.

Can I use systemctl and curl to verify my RAG service health after re-indexing?

Yes, you can use systemctl for a controlled service restart and curl to verify RAG service health, alongside checking logs to validate availability after the ChromaDB re-indexing completes.