short-term-investing-recommendation

name	short-term-investing-recommendation
description	Use when generating short-term swing-trade recommendations from this investing workspace. Runs the short-term backtest and recommendation workflow, reads tickers and risk from input_config.json, reads actual holdings from state/portfolio_state.json, and returns BUY, SELL, or HOLD ideas with technical rationale and backtest context.

Short-Term Investing Recommendation

Use this skill for the swing-trading workflow in this repo.

Working directory

Assume the skill runs from this directory:

/Users/anuragmohan/Documents/POC/investing

If the runtime starts elsewhere, change into that directory first or use absolute paths.

What this skill does

• Runs the short-term backtest and recommendation engine
• Uses the watchlist from input_config.json
• Uses real portfolio state from state/portfolio_state.json
• Produces short-term BUY, SELL, and HOLD recommendations
• Includes backtest context before trusting live recommendations

Commands

Run from the working directory above:

python3 /Users/anuragmohan/Documents/POC/investing/main.py backtest --format text
python3 /Users/anuragmohan/Documents/POC/investing/main.py recommend --format text
python3 /Users/anuragmohan/Documents/POC/investing/main.py run-all --format text

If machine-readable output is needed:

python3 /Users/anuragmohan/Documents/POC/investing/main.py run-all --format json --output /Users/anuragmohan/Documents/POC/investing/shortterm_output.json

Inputs

• Watchlist, starting capital, risk preference: /Users/anuragmohan/Documents/POC/investing/input_config.json
• Current holdings and cash: /Users/anuragmohan/Documents/POC/investing/state/portfolio_state.json

Before running, confirm these files exist and are valid JSON.

Scheduling guidance

Preferred recommendation times in US Eastern time:

• 10:15 AM
• 12:15 PM
• 2:15 PM

Preferred backtest time:

• Once daily after the market close, around 5:00 PM ET

Expected output

Summarize:

• strategy mode
• backtest metrics
• portfolio summary
• ranked recommendations

For each actionable recommendation, include:

• ticker
• action
• confidence
• price
• stop-loss
• take-profit
• explicit reason for the decision

If the action is BUY, explain why the setup qualifies now.

If the action is SELL, explain what broke down or what exit condition was triggered.

If all signals are weak or data is unavailable, explicitly report HOLD.

Guardrails

• Do not auto-update holdings after recommendations
• Do not assume a recommendation was executed
• If market data fetch fails, report that recommendations are not trustworthy
• If backtest mode is blocked, do not present new BUY ideas as actionable
• Never return BUY or SELL without also returning the reason for that decision

Post-trade note

Executed trades are recorded separately with:

python3 /Users/anuragmohan/Documents/POC/investing/update_portfolio.py --ticker TICKER --side BUY_OR_SELL --quantity QTY --price PRICE --executed-at YYYY-MM-DD