$ si-protocols
Cybersecurity for the Soul
Open-source tools to detect disinformation in metaphysical and spiritual content. Run locally. Transparent markers. No data collection.
Tech layer
spaCy NLP pipeline detecting vagueness patterns, authority claims, urgency/fear triggers, emotional manipulation (fear/euphoria polarity with contrast bonus), logical contradictions (opposing claims in the same text), source attribution analysis (unfalsifiable sources and unnamed authorities, offset by verifiable citations), and commitment escalation (foot-in-the-door progression). Weighted composite scoring across seven dimensions.
Heuristic layer
Probabilistic dissonance scanner. Placeholder for future biofeedback integration — bridging analytical and intuitive assessment.
Hybrid scoring
60% tech + 40% heuristic. Produces a 0–100 threat score with full breakdown of what triggered it.
Topology analysis
Extracts claims from text, classifies each along four axes (falsifiability, verifiability, domain coherence, logical dependency), and builds a layered graph of pseudo-variables, true-variables, and indeterminate claims. Produces intelligence-themed SVG visualisations or structured JSON. Three engine tiers: local rule-based (default), Claude API, and local-LLM (stub).
Quick Start
git clone https://github.com/lemur47/si-protocols.git
cd si-protocols
uv sync --all-extras
uv run si-threat-filter examples/synthetic_suspicious.txtRequires Python 3.12+ and uv. See the full quickstart guide for details.