AI/LLMs I built a free web tool to automate 5e prep (Visual Encounter Graph, Cursed Items, Shops). Looking for veteran feedback!

[Cmcccombs01]

@The Firebird Just a quick 1:00 AM update before I collapse: The monolith has been split, and your requests are officially LIVE!

1. The Split-Pane Editor: The Monster Lab now features a live-preview Markdown editor. You type on the left, the stat block updates instantly on the right.2. The Deterministic Hoard: The AI is gone from the Dragon's Hoard. It is now powered by a pure Python script rolling virtual d100s directly on the official 5e SRD loot tables. Zero hallucinations.3. The 3.5e Patch: Rebuilt the Rules Lawyer prompt. It should no longer suffer from 3.5e flashbacks!

I also completely decoupled the app into a Microservice architecture to handle the Viberank traffic spike, so the UI should feel incredibly snappy tonight. Thank you again for the masterclass in UX feedback. Let me know what you think of the new Monster Lab!

 

[Cmcccombs01]

Update: Just finished the WebSocket migration. Players can now see DM updates in 0.01s. Thanks to @The Firebird for pushing me to get the 'plumbing' right. It's officially a real-time workstation now.

 

[Cmcccombs01]

Infrastructure Update: 2024 RAW Bridge & Performance Transparency​

Hey everyone!

Quick update on the DM Co-Pilot development. Thanks to the blunt feedback here about the "senile wizard" (the AI hallucinating 3.5e rules), I just pushed Batch 15 to the live servers.

The Highlights:

1. The 2024 RAW Bridge: I’ve hard-coded a deterministic rules bridge between the app and the Oracle microservice. It now strictly enforces 2024 RAW (Rules as Written).
   
   • Test it out: Ask the Rules Lawyer about the new Exhaustion rules. It should correctly cite the -2 penalty to all D20 tests instead of the old 2014 or 3.5e versions.
2. Sub-Second Speeds & The "Turtle": To show I’m serious about scaling, I’ve added p99 Latency tracking to the Admin Dashboard. While our mean latency is a snappy 0.74s, I’m now monitoring the slowest 1% of users (the "Turtle" metric) to ensure no DM gets left behind during a session.
3. Live Unit-Economics: I’ve deployed a dynamic FinOps Ledger. It calculates real-time API spend ($0.0500 for the last 100 calls) so I can keep this tool free and sustainable as we climb the Viberank charts.

We are still pushing hard on our Product Hunt launch today, but my priority remains making this the most mechanically reliable tool for your weekend prep.

Let me know if the "Rules Lawyer" is still acting up or if the new 2024 weapon masteries are mapping correctly for you!

 

[Cmcccombs01]

Update: 2024 Rules Bridge & Universal AI Sync is Live!

Hey everyone! I wanted to close the loop on the "senile wizard" issue (the AI hallucinating older rules).

Batch 16 is officially live on the DM Co-Pilot servers.

1. Strict 2024 RAW: I’ve hard-coded the rules bridge into the backend logic. Ask the Rules Lawyer about Weapon Masteries or the new Exhaustion rules—it’s now strictly deterministic.
2. Universal AI Bridge: I’ve added an MCP (Model Context Protocol) server to the app. This means DM Co-Pilot can now act as a "Source of Truth" for other AI tools you might use.
3. Speed & Stability: I’m now tracking p99 Latency (the slowest 1%) on the Admin Dashboard to ensure that even with 450+ people online, your mid-combat queries stay snappy.

Thanks for the feedback that pushed me to move away from LLM "guesswork" and toward a professional-grade rules engine. Let me know how the new 2024 mappings feel!