cache-sentinel

Audits PWA cache consistency across MongoDB, Service Worker, and frontend.

Updated Aug 27, 2026
One-click install
npx skills add https://github.com/paulinett1508-dev/SuperCartolaManagerv5-production --skill cache-sentinel
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: cache-sentinel
Source: https://github.com/paulinett1508-dev/SuperCartolaManagerv5-production/tree/main/.claude/skills/cache-sentinel
Command: npx skills add https://github.com/paulinett1508-dev/SuperCartolaManagerv5-production --skill cache-sentinel

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

It prevents the participant PWA from serving stale or inconsistent data by proactively detecting cache divergence across Service Worker, IndexedDB, Memory, and MongoDB.

Core Features & Use Cases

  • Proactive stale detection for the participant app: checks cache coherence across all participant-side layers, ensuring old data cannot prevail over the system state.
  • Targeted audits by scope and timing: use focused modes like MongoDB-only, Service Worker-only, frontend (IndexedDB+Memory), or live-experience transitions.
  • Live and post-deploy verification: validates TTL behavior during market closed/open, partial updates, and consolidation-driven invalidations.
  • Baseline issue verification: re-checks known critical findings (e.g., SW cache version mismatch) to confirm they were fixed.

Quick Start

Run cache-sentinel in full participant scope immediately after a deploy by issuing the command /cache-sentinel --full.

Frequently Asked Questions about cache-sentinel

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

FAQPage Schema
How do I detect stale cache data in a PWA before going live?

Use the --full mode to audit all participant-side layers including Service Worker, IndexedDB, Memory, and MongoDB caches. This cross-verifies cache coherence and backend invalidation logic to catch stale data before it reaches users.

Why does my Service Worker serve outdated data after a deploy?

Your Service Worker cache versioning might mismatch the new deployment. Auditing Service Worker cache naming and versioning behavior cross-verifies frontend caches against backend MongoDB collections to ensure old data cannot prevail.

How do I validate IndexedDB and Memory TTL coherence during live market transitions?

Validate IndexedDB and Memory TTL coherence during live market transitions by checking cache invalidation paths for market open, closed, and partial update states. Use targeted audit modes like --live or --frontend to verify transition behavior.

Can I check only MongoDB cache collections for stale participant data?

Yes, you can check only MongoDB cache collections for stale participant data by running a targeted audit with the --mongo mode. This focused approach isolates backend cache collection checks from frontend Service Worker and IndexedDB layers.

What is the best way to investigate inconsistent PWA cache data reported by users?

The best way to investigate inconsistent PWA cache data reported by users is running a full participant scope audit. It cross-verifies MongoDB cache collections, Service Worker cache behavior, and frontend IndexedDB and Memory TTL coherence to identify divergence.

When should I not use a full PWA cache audit?

You should not use a full PWA cache audit when you need to isolate a specific layer during incident investigations. Instead, use focused modes like --sw for Service Worker, --frontend for IndexedDB and Memory, or --live for market transition invalidation paths.