investing: Skill for Claude Code

Detect the biggest financial and operating inflections for the company specified by the user: $ARGUMENTS

Before starting, read ../data-access.md for data access methods and ../design-system.md for formatting conventions. Follow the data access detection logic and design system throughout this skill.

Follow these steps:

1. Company Lookup

Look up the company by ticker using discover_companies. Capture:

• company_id
• latest_calendar_quarter — anchor for all period calculations below (see ../data-access.md Section 1.5)
• latest_fiscal_quarter
• Firm name for report attribution (default: "Daloopa") — see ../data-access.md Section 4.5

2. Broad Series Discovery

Cast a wide net to discover ALL available series for this company. Search with multiple keyword sets to maximize coverage:

• Financial: "revenue", "income", "profit", "margin", "eps", "cash flow"
• Operating: "subscriber", "user", "customer", "unit", "arpu", "retention"
• Segment: "segment", "product", "service", "geographic"
• Balance sheet: "debt", "asset", "equity", "cash"
• Other: "backlog", "bookings", "pipeline", "store", "employee"

Collect all unique series IDs. The goal is comprehensiveness — capture every metric Daloopa tracks for this company.

3. Pull 8 Quarters of Data

Calculate 8 quarters backward from latest_calendar_quarter. Pull all discovered series for those periods. This gives enough history to compute both QoQ and YoY rates plus their second derivatives.

4. Compute Growth Rates and Inflections

For each series with sufficient data (at least 5 quarters):

YoY Growth Rate for each quarter:

• growth_t = (value_t - value_{t-4}) / |value_{t-4}|

YoY Acceleration (second derivative):

• accel_t = growth_t - growth_{t-1}
• Positive = accelerating, Negative = decelerating

QoQ Sequential Growth (for non-seasonal metrics):

• seq_growth_t = (value_t - value_{t-1}) / |value_{t-1}|

QoQ Acceleration:

• seq_accel_t = seq_growth_t - seq_growth_{t-1}

Skip series where values are too small (< 1% of revenue) or where data is sparse. For margin/ratio series (values between 0-1 or percentages), compute change in basis points rather than % change.