Zero Practice Dashboard

Local web dashboard for tracking real MPK zero-cycle runs over time.

It watches your configured Minecraft instance, ingests run data into SQLite, and gives live analytics + practice targeting.

• Continuously reads your instance latest.log
• Parses run outcomes (success/fail/flyaway), tower/side/rotation, O-level, standing Y, explosions, shots, and damage splits
• Stores both raw log history and parsed attempts in data/zero_cycles.db
• Shows live charts, recent attempts, and a tower x O-level heatmap
• Recommends what to practice next and can drive set-seed practice automatically
• Tracks stronghold navigation raw path (Eye Spy -> End entry) in storage + SQLite for later room analysis

The app can inject helper components into your selected instance, with backups and restore on uninject/shutdown.

• Always injects: zdash_tracker datapack
• Optional: altered Atum jar (set-seed injection), recipe-book jar, dragon-node patch jar
• Supports Full Random override (no forced seed injection)
• Supports Legal Ranked Instance mode:
  • Forces full random
  • Disables recipe-book + dragon patch jars
  • Skips Atum injection (no set-seed injection)
  • Still injects datapack

1. Download ZIP: click green Code -> Download ZIP, then extract.
2. Open terminal in the extracted folder.
3. Install deps:

py -3 -m pip install -r requirements.txt

4. Run:

py -3 run_dashboard.py

5. Open http://127.0.0.1:8000

Alternative: run run_zero_tracker.bat to auto-clone/update, install requirements, and start.

1. Enter your MultiMC instance path (.minecraft folder).
2. Choose toggles (or Legal Ranked mode).
3. Click Save & Inject.
4. Use Uninject + Clear Path to restore and clear config. ( will also happen on application exit )

• Database: data/zero_cycles.db
• App config defaults: config.py
• Main runtime path is stored in DB (setup.mpk_instance_path)

This repo now includes a seed-based stronghold room/path cracker for 1.16.1.

1. Ensure your datapack-injected run contains Eye Spy and End entry data.
2. Run:

py -3 scripts/analyze_stronghold_world.py "C:\path\to\world"

Optional: also write rooms_entered + starter dwell back into latest DB attempt for that world:

py -3 scripts/analyze_stronghold_world.py "C:\path\to\world" --update-db

3. Output map JSON is written to:

• <world>\data\zdash_stronghold_map.json

Raw stronghold samples are also stored in DB table:

• stronghold_samples (linked to attempts.id)

Auto-watch latest world on exit (no manual world-name lookup):

py -3 scripts/watch_stronghold_analyze.py --instance "C:\Users\Boyen\Desktop\MultiMC\instances\Ranked\.minecraft"