Monitor live data and news, run dual-model valuation with an intelligent tiebreaker, create Bear/Base/Bull targets and risk scores, log results to Google Sheets, and notify via Telegram.
This AI agent automates an end-to-end institutional stock valuation workflow. It ingests live data, runs parallel analyses with two leading models, and applies an intelligent tiebreaker to reconcile disagreements. The output includes Bear/Base/Bull price targets, Buy/Hold/Sell verdicts, and a risk-adjusted confidence score, stored in Google Sheets and delivered via Telegram.
Core actions the AI agent performs to generate actionable signals.
Ingests live financial data, balance sheet, income statement, cash flow, company profile, and real-time price.
Runs parallel valuation prompts on ChatGPT and Gemini to generate independent outputs.
Triggers the intelligent tiebreaker when outputs disagree or base targets diverge by more than 20%, and averages results.
Calculates Bear/Base/Bull price targets using sector-aware multiples, growth phase recognition, P/E sanity checks, and discount-rate tiering.
Computes a confidence index from F-Score, model agreement gap, sentiment, and financial health.
Writes results to Google Sheets and delivers formatted summaries via Telegram.
This section translates the workflow into practical reasons to adopt it in real-world portfolios.
A simple 3-step flow that non-technical users can follow.
Ingests live data, news, and fundamentals, then runs parallel valuation prompts on ChatGPT and Gemini, producing independent outputs.
If outputs diverge or the base-target gap exceeds 20%, triggers the tiebreaker where bull and bear cases are averaged for the final result.
Computes Bear/Base/Bull targets and confidence scores, upserts results to Google Sheets, and sends formatted summaries via Telegram.
One realistic scenario showing inputs, process, and outcomes.
Scenario: Nightly batch across 15 tickers; 2:00 AM UTC; agent fetches data, runs dual-model valuation, applies the tiebreaker if needed, and upserts a row per ticker with targets, verdict, and confidence. If a new Buy signal emerges, a Telegram notification is sent; all results are stored with date stamps in Google Sheets.
One supporting sentence.
Automates nightly valuation and risk-aware targets, enabling faster decision-making.
Provides structured outputs that support consensus and validation.
Exposes valuation outputs via API or webhook for dashboards and apps.
Gives access to institutional-grade structure for informed decisions.
Supports backtesting and scenario analysis with historical logs.
Monitors risk-adjusted signals and confidence thresholds across portfolios.
One supporting sentence with short explanation.
Fetches balance sheet, income statement, cash flow, company profile, and real-time price data.
Ingests news sentiment feeds and feeds qualitative context into valuation prompts.
Runs parallel valuation prompts to generate bull and bear perspectives.
Provides independent dual-model valuation outputs for comparison.
Reads the watchlist and upserts per-ticker results with full history.
Delivers formatted summaries and alerts after batch cycles.
One supporting sentence with short explanation.
One supporting sentence with short explanation.
The AI agent ingests live Alpha Vantage data (balance sheet, income statement, cash flow, company profile, price) and Seeking Alpha news feeds, plus optional fundamentals. It uses internal caching to minimize repeated calls and maintains an auditable log in Google Sheets. The models (ChatGPT and Gemini) are used in tandem to form a consensus signal, with a tiebreaker applying when disagreements occur. The outputs include price targets, verdicts, and confidence scores, delivered to Sheets and Telegram.
Targets are updated on a nightly batch cycle. Each run fetches fresh data, re-evaluates with both models, and applies the tiebreaker if needed. Results are written to Google Sheets with a date stamp for historical tracking. If a new signal appears, it is surfaced via Telegram after the batch completes.
Yes. When models disagree or the base price target gap exceeds 20%, the tiebreaker assigns the bull case to one model and the bear case to the other, then averages the two outcomes. This ensures a reproducible final output rather than a coin-flip result. The final Bear/Base/Bull targets and confidence score are deterministic given the inputs and thresholds. All steps are logged for auditability.
Thresholds for triggering the tiebreaker and for base-target gaps can be adjusted in configuration. This allows tuning for different portfolios or risk tolerances. Custom thresholds are designed to preserve the structured workflow while accommodating specific governance rules. Any changes are versioned and logged with the corresponding batch date.
The confidence score combines F-Score from fundamentals, the model agreement gap, sentiment signals from Seeking Alpha, and overall financial health indicators. Each component contributes a weighted score that aggregates into a 20–90 scale. The final index reflects conviction and reliability of the three-pronged output. The score is stored alongside targets and can trigger alerts when it moves outside predefined ranges.
Results are stored in Google Sheets with per-date historical logs for each ticker. Access is controlled via Google OAuth and API permissions. Data caching reduces repeated API calls and conserves quotas, while ensuring freshness when fundamentals change. All processing happens within the configured integration stack with audit trails for governance.
Yes. The agent can expose results through webhooks or an API endpoint for downstream dashboards or research tools. Output payloads are designed as structured records including ticker, targets, verdict, and confidence. Authentication and rate limits follow standard API practices to ensure secure integration. This enables seamless incorporation into external workflows and data pipelines.
Monitor live data and news, run dual-model valuation with an intelligent tiebreaker, create Bear/Base/Bull targets and risk scores, log results to Google Sheets, and notify via Telegram.
