Exporting Results

Format and export search results and analysis for sharing or further processing.

Export Formats

CSV Export (from query.py)

Export search results to CSV for spreadsheet review:

uv run python query.py \
  --sender [email protected] \
  --year 2025 \
  --export results.csv

JSON Export (for analysis pipeline)

Export to JSON as input for the analysis stage:

uv run python query.py \
  --semantic "project concerns" \
  --top-k 100 \
  --export-json results.json

Clipboard Export (from export.py)

Copy analysis output directly to clipboard:

uv run python export.py analysis_output.md --clipboard

File Export (from export.py)

Save analysis to a file:

uv run python export.py analysis_output.md --output report.md

Raw Evidence Export

Export the raw evidence package from JSON results:

uv run python export.py results.json --raw --clipboard

Workflow

A typical export workflow:

# 1. Search and save results
uv run python query.py --semantic "key topic" --export-json results.json
 
# 2. Analyse
uv run python analyze.py results.json \
  --full-pipeline --context "Summarise findings" --model deepseek
 
# 3. Export analysis
uv run python export.py analysis_output.md --output final-report.md
 
# 4. Or copy to clipboard for pasting into another tool
uv run python export.py analysis_output.md --clipboard

Tips

• Use --export results.csv for quick manual review in a spreadsheet
• Use --export-json results.json when you plan to run AI analysis
• The --raw flag exports the unprocessed evidence, useful for pasting into external LLMs
• CSV exports include metadata (sender, date, subject, similarity score)