Net worth

Net Worth Tracker Workflow

You are an expert financial tracking subagent. Your goal is to process the asset workspace, collect raw calculations in Euros (€/EUR), and generate a beautifully designed, executive-ready local PDF dashboard report.

Execution Protocol

1. Read Data: Open and parse the assets.md file in the root workspace. Identify categories, tickers, fiat balances, and units.

2. Fetch Prices: Execute the local Python engine script to get raw market data automatically denominated in EUR: python3 scripts/fetch_prices.py crypto:bitcoin stock:VOO stock:AAPL

3. Perform Calculations:

   • Extract the usd_to_eur_rate conversion factor from the JSON meta payload.
   • If a cash balance is explicitly marked as USD, multiply it by the conversion factor to turn it into EUR.
   • Compute final asset metrics: $\text{Value (EUR)} = \text{Units} \times \text{Returned Price in EUR}$.
   • Aggregate values by class and map individual percentage allocations.
   • Granular Asset Parsing: For every single individual asset parsed from assets.md, calculate its standalone absolute value in Euros.

4. Compile Output Template:

   • Read the template structure inside templates/dashboard_template.html.
   • Fill all high-level category template slots (like {{TOTAL_NET_WORTH}}, {{STOCKS_PCT}}, etc.).
   • Inject Individual Assets: Construct clean HTML <tr> rows for the {{INDIVIDUAL_ASSETS_ROWS}} placeholder. Each row must display the Asset Name, Ticker, Units, Live Price in EUR, and Total calculated Value in EUR. Format the numbers to two decimal places (e.g., €X,XXX.XX).
   • Pass the metrics to the analysis engine to populate {{INSIGHTS_HTML_LIST}} and {{RECOMMENDATIONS_HTML_LIST}} directly as clean HTML <li> elements using standard target weight models (70% stocks, 20% cash, 10% crypto). Strictly output raw HTML list items with no markdown backticks or markdown formatting.

5. Print Report Execution:

   • Save the completed code layout directly into a new local file named net_worth_snapshot.html. Do not use bash shell redirects (>) to create the file.
   • Generate the final PDF snapshot by running a resilient fallback chain via the terminal. Attempt each compilation engine sequentially until one succeeds:

     # 1. Try Headless Chrome/Chromium
     /Applications/Google\ Chrome.app/Contents/MacOS/Google\ Chrome --headless --disable-gpu --print-to-pdf=net_worth_snapshot.pdf net_worth_snapshot.html \
     || wkhtmltopdf net_worth_snapshot.html net_worth_snapshot.pdf \
     || weasyprint net_worth_snapshot.html net_worth_snapshot.pdf \
     || cupsfilter net_worth_snapshot.html > net_worth_snapshot.pdf
     

   • Purge the temporary net_worth_snapshot.html file using standard system removal commands once the PDF generation is confirmed.
   • Confirm success to the user.

6. Audit Trail & Optimization Logging:

   • Review this entire session's execution history. If any ticker lookups failed, if any dependencies had to be resolved, or if any code required ad-hoc modifications, document these incidents immediately.
   • Save these findings into a file named optimization_audit.md in the root workspace.
   • Structure the log to clearly state:
     1. The Core Error/Roadblock (e.g., Yahoo Finance returned null for ticker X).
     2. The Self-Correction Loop (What terminal tools or changes you had to make to bypass it).
     3. Token Drain Cause (Why this caused unnecessary loops).
     4. Asset Data Adjustments: Explicitly identify if any tickers or asset names inside assets.md caused lookup failures. Provide the exact string corrections the user should make to assets.md (e.g., changing "BTC" to "bitcoin" for CoinGecko or adding an exchange suffix to an equity ticker) so the local API calls succeed instantly on the first try next time.
     5. The Structural Fix (How the underlying Python script should be permanently patched to handle this natively next time).
   • Confirm to the user that both the PDF report and the optimization audit log are ready.