# GraphRAG Builder — Python dependencies
# Install with: pip install -r requirements.txt --break-system-packages

# --- M1: Crawler ---
requests>=2.31.0
beautifulsoup4>=4.12.0
lxml>=4.9.0
playwright>=1.40.0          # Playwright install chromium also required (run once)

# --- M2: Content Extraction ---
html2text>=2020.1.16
youtube-transcript-api>=0.6.2
pdfminer.six>=20221105

# --- M3: Graph ---
networkx>=3.2.0

# --- M4: Embeddings ---
# Note: chromadb is NOT used — pure numpy index, no database dependency.
# numpy is already a transitive dep of sentence-transformers.
numpy>=1.24.0
sentence-transformers>=2.3.0

# --- M5: MCP Server ---
mcp>=1.0.0
anthropic>=0.40.0           # For concept extraction (M2 Phase 3)
