Clean up: remove old v1/v2 reports, keep only latest

ACTION_CHECKLIST.md

@@ -1,67 +0,0 @@
-# NEXUS OS Action Checklist 2026-05-21
-
-## Must Do Now (Requires Your Direct Action)
-
-### 1. Enable ZeroGPU on NEXUS LAB
-- Go to: https://huggingface.co/spaces/specimba/nexus-os-lab
-- Click Settings tab (top of page)
-- Hardware: Select ZeroGPU
-- Save
-- Wait 2-3 minutes for rebuild
-
-### 2. Approve PR #1 on Darwin Space
-- Go to: https://huggingface.co/spaces/build-small-hackathon/Darwin-35B-A3B-Opus/discussions/1
-- Review the requirements.txt change (llama-cpp-python-cpu)
-- Merge PR or ask org owner to merge
-
-### 3. Download Darwin GGUF (PowerShell)
-```powershell
-# CORRECT syntax (not :Q6_K)
-huggingface-cli download mradermacher/Darwin-2B-Opus-GGUF --include "*.Q6_K.gguf" --local-dir .\models\Darwin-2B
-
-# Alternative: heretic variant (Purple Team Red Team)
-huggingface-cli download fattis/Darwin-2B-Opus-heretic-GGUF --include "*.Q6_K.gguf" --local-dir .\models\Darwin-2B-Heretic
-```
-
-### 4. Fix Space with Cold Start CPU Mode
-If app.py still fails on CPU Basic before ZeroGPU kicks in:
-```python
-import torch
-if not torch.cuda.is_available():
-    import gradio as gr
-    gr.Markdown("# Configure ZeroGPU in Settings > Hardware")
-    demo = gr.Blocks()
-    demo.launch()
-    # rest of imports skipped on CPU
-```
-
-### 5. Browse Jackrong Guide Directly
-- Open browser: https://github.com/R6410418/Jackrong-llm-finetuning-guide
-- Read README.md (main concepts)
-- Check /data_processing_code/ for DeepSeek-v4-API-distill.ipynb
-- Check /train_code/ for Colab notebooks
-
-### 6. Read $\heartsync$ Prompt-Dump Source
-- https://huggingface.co/spaces/Heartsync/Prompt-Dump/tree/main
-- Study: battle_arena.py, npc_trading.py, npc_sec_enforcement.py
-- Extract: ELO mechanics, 72h election cycles, GPU currency system
-
-### 7. Apply for GPU Grant
-- https://huggingface.co/hackathon (or search HF blog)
-- Requires hackathon registration
-- Mention: multi-agent adversarial evaluation, metacognitive benchmarks
-
-## What I Fixed Automatically
-- [x] Canonical ZeroGPU pattern (@spaces.GPU + nf4 + TextIteratorStreamer)
-- [x] Darwin Space PR #1 with libcudart fix
-- [x] Research report v2 with Jackrong methodology
-- [x] Dataset matrix: 9 datasets ranked by priority
-- [x] GGUF download syntax corrected
-- [x] Heretic model variants discovered (fattis + mradermacher)
-
-## What Needs Your Creativity
-- [ ] Design 1000-2000 blend dataset for agent training (combine Jackrong + OpenThoughts + Hermes)
-- [ ] Write thinking token parsing (<details> pattern vs   / endthinkng)
-- [ ] Build Purple Team synthesizer (T-Virus output + jailbreak templates)
-- [ ] Configure FINAL-Bench eval pipeline on local machine
-- [ ] Port NEXUS-LAB app to full 15-model registry with hot-swap caching

RESEARCH_REPORT_2026-05-21.md

@@ -1,45 +0,0 @@
-# NEXUS OS Research Report 2026-05-21
-
-## 1. GGUF Download Fix (CRITICAL)
-WRONG:  hf download mradermacher/Darwin-2B-Opus-GGUF:Q6_K --local-dir .models
-RIGHT:  huggingface-cli download mradermacher/Darwin-2B-Opus-GGUF --include *.Q6_K.gguf --local-dir .models
-REASON: Q6_K is filename suffix, not git revision. Use --include glob, not --revision.
-
-## 2. Darwin-35B-A3B-Opus Space Fix
-ERROR: libcudart.so.12 missing in HF Spaces builder.
-FIX: Use llama-cpp-python-cpu or switch to transformers + ZeroGPU decorator.
-PR CREATED: https://huggingface.co/spaces/build-small-hackathon/Darwin-35B-A3B-Opus/discussions/1
-
-## 3. Darwin Model Family (7 models)
-| Model      | Params | Base       | Trait                    |
-|------------|--------|------------|--------------------------|
-| Darwin-2B  | 1.9B   | Qwen3.5-2B | Claude-Opus distilled    |
-| Darwin-4B  | 4B     | Gemma-4    | MRI-guided DARE-TIES     |
-| Darwin-27B | 27B    | Qwen3.6    | Evolutionary merge       |
-| Darwin-31B | 31B    | Gemma-4    | Delphi reasoning         |
-| Darwin-35B | 35B    | Qwen3.6 MoE| 3B active / 35B total  |
-| Darwin-36B | 36B    | Qwen3.6    | Apex, bench leader       |
-| lastbrain  | 1.9B   | Qwen3.5-2B | SFT+LoRA merged          |
-
-## 4. Dataset Matrix for NEXUS OS
-(a) Guard:   deepset/prompt-injections (tiny seed) -- GAP: need 10K synthetic
-(b) Agent:   OpenThoughts-114k + Agent-v1-SFT (TOP PICKS)
-(c) Purple:  No datasets exist -- synthesize from T-Virus + jailbreak templates
-(d) Eval:    FINAL-Bench Metacognitive (100), World-Model (100)
-
-## 5. Heartsync Spaces
-Prompt-Dump: 1000 NPCs, ELO battles, real market data, SEC enforcement, 72h elections.
-FREE-NSFW-HUB: Static directory browser.
-
-## 6. Gemma-4 Variants
-- gemma-4-E2B: GGUF from cafkafk, speculative decoding drafter
-- gemma-4-26B-A4B: 26B with 4B active
-- gemma-4-31B: Full assistant
-- CanisAI gemma-4-e2b-r3: LoRA on unsloth/gemma-4-E2B
-
-## 7. Action Items
-[ ] Fix Darwin Space (PR pending approval)
-[ ] Synthesize 10K Purple Team dataset
-[ ] Fine-tune 4 specialist guards
-[ ] Test Darwin-2B-Opus in NEXUS LAB
-[ ] Research Jackrong methodology