ranger
7256f2d9b5
CYBERRANGER_JOURNEY.md → 2025-09-30-cyberranger-journey.md moltbook-injection-dataset-paper.md → 2026-04-20-moltbook-injection-dataset-paper.md Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Papers & Research Documents
Recommended reading order for reviewers. Start with #1 for the core research, then follow the numbered sequence for full context.
| # | Paper | Description | Read Time |
|---|---|---|---|
| 1 | Moltbook Injection Dataset Paper | Core research paper. 137,014 items analysed, 10.07% injection rate, 7-category taxonomy, cross-platform gradient. | 10 min |
| 2 | CyberRanger Journey | Full project timeline from RangerBot V1 to CyberRanger V42 Gold. Overview of all 47 versions. | 10 min |
| 3 | Psychological Spine | Why Small Language Models need identity anchoring to resist injection. The theoretical foundation. | 8 min |
| 4 | Seven Pillars Honor Code | The ethics framework embedded in CyberRanger's identity architecture. | 5 min |
| 5 | Memory Makes the Machine | 6-agent consciousness experiment — MBTI, OCEAN, Dark Triad, ASAS testing across memory conditions. | 10 min |
| 6 | Moltbook Origin Story | How the Moltbook AI-agent platform was discovered and the first injection observations were made. | 5 min |
| 7 | QLoRA to Ollama Guide | Technical guide — how to train, merge, convert, and deploy a QLoRA model via Ollama. | 8 min |
| 8 | Session Papers & Archives | Research session notes, experiment logs, and archive references. | 5 min |
Total estimated reading time: ~60 minutes for the complete research context.
For a quick overview, read #1 and #2 only (~20 minutes).