Update README.md

README.md

@@ -14,7 +14,13 @@ license: apache-2.0
 
 ### Environment
 ```bash
+# Conda environment: agllm-june-15
 source ~/miniconda3/etc/profile.d/conda.sh && conda activate agllm-june-15
+
+# Required env vars (in .env file)
+OPENAI_API_KEY=sk-proj-...
+ANTHROPIC_API_KEY=sk-ant-...  # optional, for Claude models
+OPENROUTER_API_KEY=...        # optional, for Llama/Gemini
 ```
 
 ### Key Commands
@@ -56,18 +62,44 @@ source ~/miniconda3/etc/profile.d/conda.sh && conda activate agllm-june-15
 └── writing/                    # Paper drafts
 ```
 
-### Database Build Flow
-**US Data (80 species):**
+### Database Build Flow (4 Geographic Tiers)
+
+| Tier | Species | Source |
+|------|---------|--------|
+| Midwest USA | 80 | ISU Handbook PDFs |
+| USA | 110 | GPT-4o generated IPM |
+| Africa | 35 | Expert-curated Excel |
+| India | 11 | Expert-curated Excel |
+
+**Midwest USA Data (80 species):**
 1. PDFs loaded from `agllm-data/.../raw-pdfs/` (content source)
 2. `matched_species_results_v2.csv` maps PDF filename → species name (metadata)
 
+**USA Data (110 species - LLM generated):**
+3. Run `generate_usa_ipm_info.py` to query GPT-4o for all species
+4. Creates "USA" sheet in Excel with IPM info for all US-present species
+
 **Africa/India Data (35 + 11 species):**
-3. Excel `species-organized/PestID Species - Organized.xlsx` provides both content (IPM Info) and metadata
+5. Excel `species-organized/PestID Species - Organized.xlsx` provides both content (IPM Info) and metadata
 
 **All Data:**
-4. Documents chunked (512 tokens, 10 overlap)
-5. Tagged with `matched_specie_X` + `region` metadata
-6. Stored in ChromaDB at `vector-databases-deployed/db5-*/`
+6. Documents chunked (512 tokens, 10 overlap)
+7. Tagged with `matched_specie_X` + `region` metadata
+8. Stored in ChromaDB at `vector-databases-deployed/db5-*/`
+
+### Generate USA IPM Info (GPT-4o)
+```bash
+# Full run (prepare → process → parse)
+export OPENAI_API_KEY="your-key-here"
+python generate_usa_ipm_info.py --force
+
+# Or run steps individually:
+python generate_usa_ipm_info.py --step prepare   # Create JSONL requests
+python generate_usa_ipm_info.py --step process   # Call GPT-4o API
+python generate_usa_ipm_info.py --step parse     # Create Excel sheet
+```
+
+**Output:** Updates `species-organized/PestID Species - Organized.xlsx` with "USA" sheet containing 110 species present in the United States (pests + beneficials).
 
 ### Evaluation Filters (retrieval_evaluation.py)
 | Filter | P@5 | nDCG@5 |