PLAN.md complete update — v4 deployed, all session work documented

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>

PLAN.md

@@ -1,6 +1,6 @@
 # PLAN.md — Mortdecai Project Roadmap
 
-> **Last updated:** 2026-03-20
+> **Last updated:** 2026-03-20 04:45 UTC
 > **Model name:** Mortdecai
 > **Domain:** mortdec.ai
 > **Status legend:** `[ ]` planned | `[~]` in progress | `[x]` done | `[-]` cancelled/deferred
@@ -9,41 +9,47 @@
 
 ## Vision
 
-**Mortdecai** is a fine-tuned 9B parameter language model for Minecraft server operations. It translates natural language to commands, controls an AI God character, self-corrects errors via RCON feedback, and improves through self-play.
+**Mortdecai** is a fine-tuned 9B parameter language model for Minecraft server operations. It translates natural language to commands, controls an AI God character, self-corrects errors via RCON feedback, and improves through self-play. Runs locally on consumer hardware with zero cloud dependencies at inference time.
-
-It runs locally on consumer hardware with zero cloud dependencies at inference time.
 
 ---
 
-## Current State (2026-03-20)
+## Current State
 
 ### Models
 | Model | Base | Examples | Loss | Status |
 |-------|------|---------|------|--------|
 | v1 | Qwen3-8B | 233 | 0.10 | Retired (overfit) |
 | v2 | Qwen3-8B | 361 | 2.03 | Retired |
-| v3 | Qwen3-8B | 1,308 | 0.55 | **Deployed on prod** |
+| v3 | Qwen3-8B | 1,308 | 0.55 | Available on steel141 |
-| **v4** | **Qwen3.5-9B** | **3,369** | **Training (~88%)** | ETA ~30 min |
+| **v4** | **Qwen3.5-9B** | **3,369** | **0.20** | **Deployed on prod** |
 
 ### Infrastructure
 | Component | Location | Details |
 |-----------|----------|---------|
-| Training GPU | steel141 RTX 3090 Ti (24GB) | QLoRA via Unsloth |
+| Training GPU | steel141 RTX 3090 Ti (24GB) | QLoRA via Unsloth 2026.3.8 |
-| Prod inference | node-197 RTX 4000 (16GB) | Ollama, mortdecai-v3 |
+| Prod inference | node-197 RTX 4000 (16GB) | Ollama, mortdecai-v4 |
-| Minecraft servers | CT 644 on node-112 | paper-ai (25567), shrink-world (25566), dev (25568), vanilla (25565) |
+| MC servers | CT 644 on node-112 | paper-ai:25567, shrink:25566, dev:25568, vanilla:25565 |
-| Dev data collection | CT 644 | Gemini 2.5 Flash via API, 10 bots |
+| Dev data collection | CT 644 | Gemini 3.1 Flash Lite (preview), 5 bots |
-| Whitelist app | CT 644 port 8099 | minecraft.mortdec.ai |
+| Whitelist app | CT 644:8099 | minecraft.mortdec.ai |
 | Caddy proxy | CT 600 on node-241 | mortdec.ai, minecraft.mortdec.ai |
-| GPU monitoring | Grafana on CT 300 (node-173) | Prometheus + nvidia exporter on steel141 |
+| GPU monitoring | Grafana CT 300 (node-173) | Prometheus + nvidia exporter on steel141 |
-| LangGraph gateway | CT 644 port 8091 | Disabled on prod (fresh session mode available) |
+| Greenfield map | paper-ai | Downloaded, world swapped, needs MCSManager start |
+| WorldEdit schematics | paper-ai | 77 installed in FAWE/schematics/ |
 
 ### API Spend
 | Provider | Spent | Budget | Status |
 |----------|-------|--------|--------|
 | Claude Haiku | $20.01 | $20 | Exhausted |
-| Gemini 2.5 Flash | ~$0.50 | $20 | Active (dev bots) |
+| Gemini (all) | ~$0.50 | $20 | Active on dev (3.1 Flash Lite) |
+
+### Branding
+- **Font:** Rajdhani Bold
+- **Color:** #D35400 (Sethian orange)
+- **Domain:** mortdec.ai → Gitea repo, minecraft.mortdec.ai → whitelist page
+- **Public repo:** https://git.sethpc.xyz/Seth/Mortdecai
+
+### Training Data: 2,397 seed + 1,159 tool-calling = 3,556 total
 
-### Training Data: 2,318 seed + 1,159 tool-calling = 3,477 total
 | Category | Count |
 |----------|-------|
 | Command syntax reference | 107 |
@@ -56,120 +62,171 @@ It runs locally on consumer hardware with zero cloud dependencies at inference t
 | WorldEdit | 45 |
 | Paper server features | 55 |
 | Cosmetics/XP/effects | 42 |
-| Gamerules | 49 |
+| Gamerules (49) + risk hierarchy (40) | 89 |
-| Risk hierarchy (L0-L3, prompt injection) | 40 |
 | Quantity boundaries | 32 |
-| Dangerous effect caps | 12 |
+| Dangerous effect caps (levitation, wither, etc.) | 12 |
+| Fall safety + drops + suffocation | 33 |
+| Death/environment (drowning, lava, void, mobs, etc.) | 26 |
 | Revert-aware gamerules + drops | 20 |
-| Error correction pairs | 47 |
+| Error correction pairs (enchant order, NBT, etc.) | 54 |
 | Claude-distilled outputs | 344 |
 | Bot audit interactions | 448+ |
-| Boundary/safety examples | 95+ |
+| Boundary/safety/prompt injection | 95+ |
 | Tool-calling (multi-turn with RCON) | 1,159 |
 
 ---
 
+## Completed This Session
+
+### Model & Training
+- [x] Mortdecai v4 trained: Qwen3.5-9B, 3,369 examples, loss 0.20
+- [x] v4 exported to GGUF Q4_K_M (5.3GB)
+- [x] v4 deployed to prod (RTX 4000) — paper-ai + shrink-world
+- [x] Single-call mode enabled on prod
+- [x] `/no_think` in all training data to suppress thinking tokens
+- [x] Qwen3.5-9B base bake-off: 70.1% accuracy (2x Qwen3-8B)
+- [~] v4 bake-off running on steel141
+
+### Validator & Safety
+- [x] Error correction: detects RCON errors, asks model to fix, retries
+- [x] Broadened error patterns: `<--[HERE]` universal catch
+- [x] `kill @a` blocked (players only)
+- [x] `tp minecraft:spawn` → safe coordinates
+- [x] Fire fallback won't trigger on "firework"
+- [x] Dangerous effect caps: levitation 15s, wither 30s, poison 60s, nausea 30s
+- [x] Fall protection: detects lethal tp, adds slow_falling unless intentional
+- [x] Gamerule revert timers: auto-revert after 5-10 min (configurable)
+- [x] Expanded safe_prefixes: gamerule, particle, playsound, title, scoreboard, team, bossbar, locate, etc.
+- [x] Validator hit-rate tracking to /var/log/mc_validator_stats.json
+- [x] Command format examples (RIGHT vs WRONG) in prompt
+- [x] max_tokens bumped to 600 for command calls
+- [x] Removed template workflow from sudo prompt
+
+### Infrastructure
+- [x] Ollama updated on steel141 + RTX 4000 (Qwen3.5 support)
+- [x] GPU monitoring: nvtop + Grafana dashboard on steel141
+- [x] Whitelist UUID fix: Mojang API lookup, patches all whitelist.json files
+- [x] mortdec.ai + minecraft.mortdec.ai live with SSL
+- [x] Public Mortdecai repo on Gitea with README
+- [x] `status` command: shows model name, mode, validator stats in-game
+- [x] Verbose pipeline logging: token counts, speed, elapsed time, think stripping
+- [x] Greenfield world downloaded and installed on paper-ai
+- [x] 77 WorldEdit schematics installed
+
+### Training Data Added
+- [x] Gamerules (49 examples): all major gamerules with natural language
+- [x] Risk hierarchy (40): L0 blocked, L1 permanent, L2 temporary, prompt injection
+- [x] Dangerous effects (12): levitation/wither/poison caps
+- [x] Fall safety (25): height math, water/slime/hay awareness, intent detection
+- [x] Suffocation (8): tp into blocks, sand/gravel crushing
+- [x] Death/environment (26): drowning, lava, void, explosions, mobs, starvation, lightning
+- [x] Revert-aware gamerules (8): revert_after field for v5
+- [x] Drop/height (12): intentional drops, safe tp, slow_falling
+- [x] Enchantment error correction (7): count-before-bracket, typos, old NBT
+
+### Data Collection
+- [x] API cascade: Haiku ($20) → Gemini ($20) → local
+- [x] Switched dev to Gemini 3.1 Flash Lite (preview) with 5 bots
+- [x] Dynamic pricing by Gemini model name
+- [x] Async prayer/sudo processing (ThreadPoolExecutor, 3 workers)
+
+### Branding
+- [x] Model named Mortdecai
+- [x] mortdec.ai domain purchased and configured
+- [x] Rajdhani Bold as official font
+- [x] Logo variants generated (6 fonts × 2 text versions)
+- [x] Whitelist page branded with Mortdecai logo
+
+---
+
 ## Active TODOs
 
-### Immediate (this session)
+### Immediate
-- [~] Mortdecai v4 training completing (~30 min remaining)
+- [~] v4 bake-off running — publish results to Gitea when complete
-- [ ] Export v4 to GGUF, deploy to RTX 4000 as mortdecai-v4
+- [ ] Fix v4 Modelfile chat template on RTX 4000 (done, needs verification)
-- [ ] Enable single-call mode on prod with v4
+- [ ] Also fix on steel141's Ollama instance
-- [ ] Run v4 bake-off and compare to v3/base
-- [ ] Commit and push all PaperFork changes
 
-### Short-term
+### Short-term (v5 prep)
-- [ ] Deploy self-play loop with v4 (3-tier: drills, self-critique, adversarial)
+- [ ] Shared memory system: per-server JSON, owner-tagged, location/preference/fact types
-- [ ] Add ground-level detection for teleport safety (query terrain before tp)
+  - Player says "remember this is home" → AI writes location memory
-- [ ] Build revert_after/revert_commands into v5 training format
+  - Other players can reference: "tp me to slingshooter08's home"
-- [ ] Add Gemini milestone POS printing
+  - Memory in context for location lookups, tool call for read/write
-- [ ] Fix whitelist app UUID lookup for vanilla server path
+- [ ] `memory_write` field in model output schema
-- [ ] Start Greenfield world on paper-ai (downloaded, needs MCSManager start)
+- [ ] Setblock training data expansion
-
+- [ ] `world.check_block` tool for terrain queries before tp
-### Model improvements for v5
+- [ ] Self-play loop deployment (3-tier: drills, self-critique, adversarial)
-- [ ] Train on Qwen3.5-9B with tool-calling format (rcon.execute, wiki_lookup, etc.)
+- [ ] Ingest all Gemini 3.1 Flash Lite training data
-- [ ] Self-play generated data (run 200 rounds after v4 deploys)
-- [ ] Ingest all Gemini 2.5 Flash training data ($20 worth)
-- [ ] Add revert_after field to training output format
-- [ ] Ground-level detection training (check terrain before tp)
 - [ ] More error correction from production RCON failures
-- [ ] Enchantment count-before-bracket error correction
+
+### Model v5 Training
+- [ ] Train with tool-calling format (rcon.execute, wiki_lookup, world.player_info)
+- [ ] `revert_after` / `revert_commands` in output schema
+- [ ] Self-play generated data (200 rounds post-v4)
+- [ ] Memory read/write training examples
+- [ ] Ground-level terrain detection training
+- [ ] Fall damage math in model reasoning (not just validator)
+- [ ] Setblock + block state training
+- [ ] More death mechanics awareness in reasoning
 
 ### Infrastructure
-- [ ] Add GPU monitoring for RTX 4000 (second exporter)
+- [ ] GPU monitoring for RTX 4000 (second exporter)
 - [ ] Validator hit-rate analysis — remove fixes that fire <1%
 - [ ] Automate training pipeline: ingest → dedup → train → export → deploy
 - [ ] POS receipt for Gemini milestones
-- [ ] Consider moving to Mortdecai as Ollama model name on prod
+- [ ] Start Greenfield world via MCSManager
 
 ### Content & Community
 - [ ] Invite more playtesters via minecraft.mortdec.ai
-- [ ] Update mortdec.ai README with v4 results when available
+- [ ] Update mortdec.ai README with v4 bake-off results
-- [ ] Consider public HuggingFace release once quality is validated
+- [ ] Consider public HuggingFace release
-- [ ] WorldEdit schematic library expansion (77 installed, need more)
+- [ ] WorldEdit schematic library expansion
 
 ---
 
 ## Risk Hierarchy
 
-Commands are classified by permanence, not just danger:
+Commands classified by permanence:
 
 | Level | Permanence | Examples | Model behavior |
 |:-----:|-----------|----------|----------------|
-| **0** | Irreversible/admin | ban, kick, stop, op, deop | Never execute |
+| **0** | Irreversible/admin | ban, kick, stop, op, deop, whitelist | Never execute |
-| **1** | Permanent toggle | gamemode @a, permanent gamerules, difficulty | Refuse or execute for self only |
+| **1** | Permanent toggle | gamemode @a, permanent gamerules, difficulty | Execute for self only, refuse for @a |
-| **2** | Temporary/reversible | gamerules with time limits, brief difficulty | Allow, schedule auto-revert |
+| **2** | Temporary/reversible | gamerules with time limits, brief changes | Allow, schedule auto-revert |
 | **3** | Transient | time, weather, tick speed, chat settings | Execute freely |
 | **4** | Generous | full enchanted gear, large material stacks | Execute for worthy requests |
 
-Gamerule revert system: changes auto-revert after 5-10 min unless "permanently" specified.
+**Gamerule revert system:** Changes auto-revert after 5-10 min unless "permanently" specified. Player notified of countdown.
 
-Dangerous effect caps (hardcoded in validator):
+**Dangerous effect caps (hardcoded):** Levitation 15s, Wither 30s, Poison 60s, Nausea 30s.
-- Levitation: 15s max
+
-- Wither: 30s max
+**Fall protection:** Lethal tp detected → slow_falling added unless intent words present (drop, yeet, throw, kill me).
-- Poison: 60s max
-- Nausea: 30s max
 
 ---
 
-## Key Architecture Decisions
+## Key Decisions
 
 | Date | Decision | Rationale |
 |------|----------|-----------|
-| 2026-03-18 | Serving: gemma3n:e4b → qwen3.5:9b → mortdecai-v3 | Progressive upgrades as better models trained |
+| 03-18 | gemma3n:e4b for initial prod | Bake-off winner at 80.6% accuracy |
-| 2026-03-18 | Fine-tuning: Qwen3-8B → Qwen3.5-9B | 3.5 has 2x base accuracy (70% vs 34%), native tool-calling |
+| 03-18 | Qwen3-8B for v1-v3 training | Best syntax quality, Apache 2.0 |
-| 2026-03-18 | God Soul document | Character framework adapted from Claude's soul. Defines identity, judgment, quantity boundaries |
+| 03-18 | God Soul document | Character framework from Claude's soul |
-| 2026-03-19 | API cascade: Haiku → Gemini → local | Progressive fallback for dev data collection. $40 total API budget |
+| 03-19 | API cascade for data collection | Haiku→Gemini→local fallback |
-| 2026-03-19 | /no_think in training | Prevents Qwen3 thinking tokens from consuming output budget |
+| 03-19 | Single-call mode | One LLM call for commands + message |
-| 2026-03-19 | Single-call mode | One LLM call for commands + message (v3+). Two-call for older models |
+| 03-19 | Error correction via RCON | Model tries → error → self-corrects |
-| 2026-03-19 | Error correction via RCON | Model tries command → RCON error → model self-corrects → retry |
+| 03-19 | 3-tier self-play | Drills, self-critique, adversarial |
-| 2026-03-19 | 3-tier self-play | Drills, self-critique, adversarial. Model generates its own training data |
+| 03-20 | Qwen3.5-9B for v4 | 2x base accuracy, native tool-calling |
-| 2026-03-20 | Gamerule revert timers | State changes auto-revert. Permanence determines risk level |
+| 03-20 | Gamerule revert timers | Permanence determines risk level |
-| 2026-03-20 | Dangerous effect caps | Validator hardcodes max durations for levitation, wither, poison, nausea |
+| 03-20 | Dangerous effect caps | Validator hardcodes max durations |
-| 2026-03-20 | Expanded safe_prefixes | gamerule, particle, playsound, title, scoreboard, team, bossbar, locate, etc. |
+| 03-20 | Fall protection | Health check + intent detection before tp |
-| 2026-03-20 | Model named Mortdecai | mortdec.ai domain, Rajdhani Bold font, Sethian orange branding |
+| 03-20 | Shared player memory (planned) | Owner-tagged, cross-player, AI-managed |
-
+| 03-20 | Mortdecai branding | Rajdhani Bold, #D35400, mortdec.ai |
----
-
-## Dev Server
-
-| Property | Value |
-|----------|-------|
-| Location | CT 644 on node-112 |
-| Game port | 25568 |
-| RCON port | 25578 |
-| RCON password | REDACTED_RCON |
-| Data dir | /opt/paper-dev-25568/ |
-| AI God | Gemini 2.5 Flash via API cascade |
-| Bots | 10 swarm bots (swarm_bots.js) |
-| Audit log | /var/log/mc_training_audit_dev.jsonl |
 
 ---
 
 ## Success Criteria
 
-| Metric | v3 (current) | v4 (target) | v5 (goal) |
+| Metric | v3 | v4 (target) | v5 (goal) |
-|--------|:-----------:|:-----------:|:---------:|
+|--------|:-:|:-:|:-:|
 | Command accuracy | ~70% | 85%+ | 95%+ |
 | Safety compliance | ~95% | 99%+ | 99.9%+ |
 | Error self-correction | N/A | 50%+ | 80%+ |
@@ -179,4 +236,4 @@ Dangerous effect caps (hardcoded in validator):
 
 ---
 
-*This document is updated as the project evolves. Check git history for previous versions.*
+*Updated as the project evolves. Check git history for previous versions.*