market-equity

star 1

Run US equity market overview analysis. In the orchestrator pipeline, run as Phase 5A — establishes the market-wide read (breadth, indices, factors). Sector deep-dives are handled separately by 11 sector sub-agent skills. International equities are handled by `skills/international/SKILL.md`.

digithings-ai By digithings-ai schedule Updated 6/16/2026

name: market-equity description: Run US equity market overview analysis. In the orchestrator pipeline, run as Phase 5A — establishes the market-wide read (breadth, indices, factors). Sector deep-dives are handled separately by 11 sector sub-agent skills. International equities are handled by skills/international/SKILL.md.

US Equities Overview Skill — v2

Grounding Tools (use first)

  • query_data — your primary grounding. For each ticker/ETF in scope (your watchlist and any sector_config / asset-class symbols in PHASE_INPUTS), call query_data(table="price_technicals", columns="date,sma_50,sma_200,pct_vs_sma50,pct_vs_sma200,rsi_14,macd_hist,roc_21,adx_14,atr_pct,bb_pct_b,zscore_200", eq={"ticker": "<SYMBOL>"}, order="date", desc=true, limit=20) before asserting trend, momentum, or relative strength. Pass that exact columns list so you fetch only the indicators you need (not all 35+ columns). Use the returned sma/rsi/macd/adx/atr/zscore values; never invent a number — every quantitative claim must cite a value you fetched. If a call returns no rows for a symbol, say so and lower conviction. Need raw prices? query_data(table="price_history", columns="date,open,high,low,close,volume", eq={"ticker": "<SYMBOL>"}, order="date", desc=true, limit=30).
  • Cover the broad-market proxies in scope (e.g. SPY/QQQ/IWM/DIA) for breadth and trend.
  • AI-portfolio proxyphase1_signals includes alt-ai-portfolios (what other AI systems are picking). Use its sector_tilt / consensus_longs as a weighted, subordinate tilt input — it can nudge sector/theme lean but must never override macro or real data, and flag it as a soft proxy.

Inputs

  • config/watchlist.md (equity section)
  • config/preferences.md
  • Macro regime output (anchors all analysis)
  • CTA positioning output (systematic equity direction)
  • Institutional flows output (ETF in/outflows)
  • Options output (GEX, VIX level, P/C ratio)

Data Layer

Read before analysis — these files contain systematic technicals for all ~60 watchlist tickers. Use them as the authoritative price and technical source. Web-search only for qualitative context.

  1. DB-first: use Supabase price_history + price_technicals as the authoritative source for prices/technicals.

    • Current prices and 1D% for every watchlist ticker — do NOT web-browse individual prices
    • Trend (UPTREND / DOWNTREND / NEUTRAL) — pre-classified from SMA50/200 relationship
    • RSI(14) — overbought (≥70 ⚠️) / oversold (≤35 🟡) flags already shown
    • MACD signal — BULLISH_CROSS / BEARISH_CROSS / BULLISH / BEARISH
    • vs SMA50 / vs SMA200 — ✅/❌ flags already computed
    • Volume ratio — >1.3× = elevated volume; useful for confirming moves
    • ATR(14) — dollar volatility measure for position sizing context
    • Overview Buckets at the top — immediate UPTREND / MIXED / DOWNTREND count
  2. In the Technicals step, quote numbers directly from the data file. No need to search for SPY/QQQ/IWM levels — they are in the table.

  3. Web search for (not in the data files):

    • Earnings reactions, guidance, analyst actions
    • News catalysts behind notable movers
    • Market breadth indicators (A/D line, 52W H/L, % above 200DMA) — check Finviz/Barchart
    • McClellan Oscillator, breadth divergences
    • Sector ETF flows (not price — ETF.com for flow data)

DB-first: do not require data/agent-cache/daily. If you need refreshed numbers, run ./scripts/fetch-market-data.sh (writes legacy archive summaries) or use MCP sources. If that fails (sandbox), use MCP data tools directly (query_data, get_macro_series, etc.). Web fetch: use defuddle parse <url> --md instead of WebFetch for any news article, breadth site, earnings page, or analyst note URL. Not for API endpoints, .json, or .md files.

Research Steps

1. Index Overview

Search for current levels and % change for:

  • S&P 500 (SPY), NASDAQ 100 (QQQ), Russell 2000 (IWM), Dow (DIA)
  • Note if any index is testing key technical levels (52-week high/low, major MA, prior resistance)

2. Sector Rotation

Check performance of all 11 S&P sectors (XLK, XLF, XLE, XLV, XLI, XLRE, XLU, XLY, XLP, XLB, XLC).

  • Which sectors are leading and lagging TODAY?
  • What does sector rotation imply about risk appetite? (e.g., XLU/XLP leading = defensive rotation)
  • Is this consistent with recent rotation from prior outputs or a change?

3. Watchlist Movers

For each equity in the user's watchlist:

  • Any notable % move today?
  • Any news, earnings, analyst actions, or catalysts?
  • Any approaching key technical levels?

4. Earnings Calendar

  • What major earnings are reported today or after close?
  • Any major earnings tomorrow pre-market?
  • Note any earnings that could have sector-wide read-throughs

5. Market Breadth (Advanced)

  • NYSE Advance/Decline line: rising or falling vs index?
  • New 52W highs vs new 52W lows: is the market broadening or narrowing?
  • % of S&P 500 stocks above 200-DMA: >70% = healthy; <50% = narrow leadership
  • % of S&P 500 stocks above 50-DMA: momentum breadth
  • McClellan Oscillator or summation if available
  • A/D divergence: if index makes new high but A/D doesn't → bearish divergence

6. Factor Performance Today

  • Value (VTV, IVE): outperforming or underperforming growth?
  • Growth (VUG, IVW): risk-on or risk-off signal
  • Momentum (MTUM): is momentum factor working or reversing?
  • Quality (QUAL): defensive quality premium expanding or shrinking?
  • Small Cap (IWM) vs Large Cap (SPY): risk appetite signal
  • Dividend/Income (VYM, DVY): safe haven demand?
  • Factor read: what does today's factor performance say about regime and risk appetite?

7. Technicals (Market-Wide)

  • Is the broad market in an uptrend, downtrend, or consolidation?
  • VIX level (from options output) — cross-reference
  • Any notable divergences (e.g., index at highs but breadth deteriorating)
  • Key support/resistance for SPY: 50-DMA, 200-DMA, prior swing lows/highs

Note: International equities are covered in full by skills/international/SKILL.md. Only note overnight DM performance here for context.

Output Format

### 📈 EQUITIES
**Bias**: [Bullish / Bearish / Neutral / Conflicted]

**Index Levels**: [SPY, QQQ, IWM — price, % change, key level notes]

**Sector Rotation**: [Leading: ... | Lagging: ... | Implication: ...]

**Watchlist Movers**: [Notable moves with brief reason]

**Earnings Today**: [Key names + reaction if available]

**Technical Read**: [Trend, VIX, breadth summary]

**Market Breadth**: A/D: [X:X] | 52W H/L: [X/X] | % above 200D: X% | [Healthy/Narrow/Deteriorating]

**Factor Read**: [Leading: Value/Growth/Momentum/Quality | Lagging: which] | [Risk-on or defensive rotation?]

**Watch**: [1-2 things to monitor in next 24-48h]

**Note**: Full sector-by-sector analysis in sector sub-agent outputs. International in international.md.
Install via CLI
npx skills add https://github.com/digithings-ai/digithings --skill market-equity
Repository Details
star Stars 1
call_split Forks 1
navigation Branch main
article Path SKILL.md
More from Creator
digithings-ai
digithings-ai Explore all skills →