Quant Garage

Trade like a pro. Without the terminal. View the full landing page →

Quant and equity research tools that run inside Claude, or behind your own UI. You ask Claude "preview NVDA earnings" or "screen for momentum names that pulled back this week" and you get back what a professional analyst would write at 6am, with the supporting numbers and citations to the API calls underneath.

Or you skip Claude entirely. pip install quant-garage and call the same tool from your own code. Every skill is an importable Python function that returns JSON, so it drops straight into a Jupyter notebook, a research dashboard, a Slack alert, a cron job. Both paths work because every skill ships the same compute as two layers: the JSON contract for developers and a rendered note, table, stream, or report for humans.

Forty-two primitives plus thirteen one-command workflows. One framework. Built in the garage, not the trading floor.

Needs a Massive API key. Free Basic tier runs twenty-one of the primitives plus most workflows end-to-end; $29/month Stocks Starter opens thirty-nine of the forty-two primitives and every workflow. One primitive (prediction-market-monitor) needs no Massive key at all — it uses Kalshi's public read-only API.

Feedback welcome. Found a bug or have an idea? Open an issue or send a pull request.

What the collection does

Each primitive is useful on its own. The point of having forty-two that share data, methodology, and audit trail is that they chain. The eight workflows show what that chaining looks like when someone wires the pieces together for a specific cadence.

Tuesday morning, you're long NVDA into Thursday's print. earnings-drilldown shows the implied move is rich vs the 8-quarter realized. valuation-sanity-check --mc puts the current price at the 78th percentile of peer-driven fair values. You like the setup but want to size honestly. position-sizer puts NVDA at 8% under vol-target given the 48% realized vol. You execute. portfolio-mark walks the snapshot fallback chain; slippage-cost flags one fill that crossed the spread. End of day: risk-report shows NVDA now drives 35% of portfolio variance.

One repo, one Massive key, one methodology.

The research side has the same shape. universe-builder screens for momentum pullbacks; factor-research confirms which factors are working this regime; news-scanner checks for catalysts on the survivors; event-study measures the abnormal return around each catalyst. You get from "what should I look at?" to "here's the regime context and the abnormal-return distribution" in a single workflow.

Each tool plugs into the same quant_garage/ core: same client, same timezone handling, same audit-trail format, same significance thresholds.

The idea

Every number in this repo cites a live API call. Every take is computed from actual readings. Every peer set carries its sample size. When the data is thin, the tool says so instead of guessing.

That's the whole difference between a research tool and a chatbot with a market-data plugin. The tools refuse to fabricate when the data isn't there, and they surface the endpoint and timestamp for each figure so you can retrace the reasoning.

The methodology references inside each skill folder are where the IP lives: statistical methods, sample-size rules, base rates, edge cases, honest caveats about what the take does and doesn't prove. The Massive API provides the inputs. Everything else is workflow.

LLMs and agents work better on top of this surface than under it. Ask Claude Code "review my book" and it invokes the composite; ask any tool-use LLM and it consumes the same JSON. The dual layer means humans read a briefing, agents read a schema, both are anchored to the same live citation trail.

Who this is for

Four honest audiences.

Retail investors with a real book. $50K to a few million. You make informed decisions, you don't day-trade, you don't have a Bloomberg terminal, and you want sell-side-quality output on the positions you actually hold. The workflows above (portfolio-review, weekly-brief, preflight-trade) were built for this exact use pattern.

Fintech and market-data developers. You're building on top of Massive (or considering it) and you need reference tooling that shows what a serious integration looks like. Every skill is a working data layer with source-cited outputs, retry logic, and proper fallback chains. Fork it, adapt it, ship it.

LLM and agent developers. You're building finance-adjacent agents that need tool-use with real citations, not fabricated numbers. Every run() returns JSON matching a schema; every skill is discoverable as a Claude Code skill under skills/. The dual- layer contract is designed for exactly this use case.

Analysts and PMs, as supplementary tooling. You have Bloomberg or Factset. You don't need this to replace them. You do need it for the scriptable workflows the terminal is bad at: batch a watchlist scan, feed the output into a Slack channel, run a sanity check that's reproducible and citation-anchored.

Not for: people looking for a trading system. These tools encode analyst workflow, not strategy. The takes are pattern- matchers grounded in methodology; they aren't a signal engine. If you want production alpha, you build on top of this.

Thirteen one-command workflows

Start here. Workflows are chains of the 26 building-block tools below, wired for specific cadences. Each takes a watchlist or a ticker (or nothing, for macro-only reads) and returns a single briefing with a headline block up top and the full per-tool detail below. If you only try one thing in this repo, try one of these.

portfolio-review The full book review in one call. Chains eight sub-skills: market- regime, sector-rotation-signal, historical-analog-finder, risk-report, earnings-blackout, macro-event-calendar, corporate-actions-scanner, portfolio-rebalancer. Headline block distills each into one line: regime, rotation theme, 90-day forward SPY distribution, portfolio vol plus top variance contributor, next earnings, next macro, top 8-K, rebalance verdict. Built after a live review missed an ALLO public offering (87.5M shares, 34% dilution) because the workflow was run manually and one tool was skipped. Now the workflow is one command that doesn't skip.

weekly-brief Sunday-night prep. Watchlist-focused, not position-focused. market- regime, sector-rotation-signal, macro-event-calendar (7-day window), earnings-blackout (7-day window). The Sunday briefing that frames the week: what's the tape, what's rotating, what prints, what macro matters.

morning-brief 60-second daily open. market-regime, macro-event-calendar (today + tomorrow), news-scanner (last N per watchlist ticker). Sharp counterpart to weekly-brief: shorter horizon, news-focused. Cron this at 8am for a briefing before you open the trading window.

preflight-trade Before you hit execute. Takes a ticker plus intended action (buy, sell, add, reduce, exit). Chains technical-briefing, earnings- blackout (14d), news-scanner (last N), corporate-actions-scanner (90d). Returns a deterministic verdict (go, wait, review) plus red and green flag lists. Not a recommendation to trade or not trade. A structured "is now obviously a bad time" gate.

earnings-week-prep For weeks where four of your names print in five days. earnings- blackout on the watchlist to find who prints, then earnings-drilldown plus technical-briefing for the top-N imminent prints. The Sunday- night briefing before a heavy earnings week.

historical-comparison Twin decision-support. event-study on the specific event plus historical-analog-finder on the market regime. Both anchors together so you're not relying on one. Analog-only mode for when you don't have a specific event to study.

scan-and-frame Regime-framed idea generation. market-regime for context, universe- builder for candidates, relative-strength to rank the top N. Optional factor-research pass for factor context (heavy — off by default). Discovery-mode, not position-mode.

stock-one-pager Retail-tier single-name card. technical-briefing plus earnings- blackout plus market-regime, translated into plain language. The thing to read before you buy something you saw on social. Every claim is gated to what the data actually says; nothing is templated.

Use cases, mapped to workflows

Concrete situations, mapped to the workflow that solves them.

For the 42 individual primitives (compose your own workflow), scroll down.

The 46 building-block tools, with real use cases

The workflows above are chains of these. If you're building your own workflow or agent, this is the shelf of primitives to compose from. Each is a standalone run() -> dict with a rendered layer for humans and a JSON layer for downstream code.

Earnings work

earnings-drilldown You're long NVDA into Thursday's print. Trim, hold, or fade the straddle? Run the tool. You get the implied move vs the 8-quarter realized average, the post-earnings drift t-stat conditional on the reaction direction, and which semis trade with NVDA on print days. Output reads like a sell-side morning note: bold take at the top, supporting numbers below.

event-study You want to measure abnormal returns around any event class: earnings, dividend changes, large volume spikes. Single event for one ticker (gets you a note), the same event across many tickers (cross-section table), or all events in a window (aggregate stats with t-stats). Last month's run on mega-cap tech Q1 prints surfaced that the cross-section average is negative despite all five beating on EPS. Guidance is dominating headlines this regime.

earnings-blackout You're running a watchlist and want a 30-second pre-trade hygiene check: which names print this week, which printed yesterday, which are clear. Run the tool. It returns each ticker bucketed into blackout-imminent (0-3 days forward), blackout-soon (4-7), just-printed (0-3 past), or clear, with the date, consensus EPS where Benzinga is wired, and the 8-K item code where it falls back to SEC EDGAR. Exception-report style: imminent blackouts on top, clear names at the bottom.

Equity research and valuation

pitch-comps You're a junior banker or an equity research analyst building a CRM comp set. You need the software peer group with multiples, growth, EBITDA margin, plus a regression-adjusted view that controls for the growth differential. Run it, get a table you can drop straight into the deck or the model. The one-sentence read at the bottom is the take the MD or PM wants on page two.

valuation-sanity-check The analyst on your team has a $250 NVDA target with assumed 28% growth and 60% margin. Does it survive a peer-distribution sanity check? The tool runs the reverse-DCF at the current price and tells you what CAGR is actually priced in. When I ran it, the $250 target came back understated against the semi peer set, which was the opposite of what I expected. Pass --mc for the Monte Carlo fan-of-outcomes mode: 10,000 samples across the peer growth × margin × exit-multiple distribution, with the current price's percentile within the resulting fair-value distribution and a per-driver sensitivity ranking. More honest than a single point estimate.

position-sizer You like the names — NVDA, AMZN, GOOGL, META all going in. How much of each? Run the tool and get four canonical sizing methods side by side: vol-target, fractional Kelly, risk parity, equal weight. The methods usually disagree; that's the point. Vol-target cuts the high-vol names so they don't dominate. Kelly tilts toward names with the highest edge per variance (you supply the edges). Risk parity equalizes each name's contribution to portfolio variance. Pick the column whose worldview matches your conviction.

Quant research and screening

technical-briefing You're staring at a single name and want the technical read before you do anything else. Run the tool on NVDA. You get the composite trend regime (bullish / bearish / neutral, strong or weak, with the reasons that drove the label), RSI 14 with bucketed momentum read, MACD line vs signal with cross status, SMAs 20 / 50 / 200, Bollinger position, ATR as a percent of price, and the ADV-bucketed liquidity context. Output is a sell-side morning-note block; the Take is computed from the actual readings, not a hardcoded narrative.

universe-builder You want every US common stock above $20 with quarterly +10% momentum that pulled back this week. Run the screen. The tool walks the full 12,000-name universe, applies your filter chain, ranks the survivors by composite z-score, and flags sector concentration. Last week's run surfaced a trucking cluster (ARCB, RXO, SNDR, TFII, WERN, SAIA) all off the same macro freight pullback. Real mean-reversion candidates.

factor-research You want to know what factor is working in the current regime. Run the tool on the S&P 500 over a 5-year window. You get the per-factor IC at 1M/3M/6M/12M horizons, t-stats, decile long-short spreads, hit rates, and the cross-factor correlation matrix. Quality at +3.1 t-stat is the only single factor with statistical significance right now. Low-vol is negative across every horizon (recent regime rewards risk-taking).

relative-strength You have a basket of names and want to know which are leading and which are bleeding vs SPY (or any benchmark) across 5 / 20 / 60 / 120-day windows. Run the tool. Each name gets RS in basis points per window, a composite percentile rank within the basket, and a trend label (stable_leader, improving, deteriorating, stable_laggard, mixed). Pass --include-sectors to add the 11 SPDR sector ETFs into the ranking and see whether NVDA's strength is the name or just XLK.

pairs-scanner You have a sector basket and want to know which two names are statistically tethered right now. Run the tool. Every pair gets an Engle-Granger cointegration test on log prices, the ADF t-stat bucketed against MacKinnon 2010 critical values, an Ornstein-Uhlenbeck half-life of the spread, the current z-score, and a 70/30 out-of-sample stability check that flags regime shifts. Tradeable pairs surface at the top (widest spreads first); the considered-but-rejected list shows why the others didn't make it. Screen, not strategy.

hurst-exponent Companion to pairs-scanner but single-name. Runs rescaled-range (R/S) analysis on 2 years of daily log returns and reports the Hurst exponent with a bootstrap confidence band. Classifies the name as mean-reverting (H < 0.45), random walk (H in [0.45, 0.55]), or trending (H > 0.55). Answers "is this name a mean-reversion setup or a momentum setup?" Utilities and staples cluster on the low side, growth names in strong runs on the high side.

change-point-detector Bayesian Online Change-Point Detection (Adams & MacKay 2007) on a ticker's daily log returns. Detects the specific dates where the return-generating distribution shifted (regime change in mean, vol, or both), reports posterior confidence at each boundary, and emits per-segment annualized return + vol so you can see what changed. Live on ALLO: the 2026-04-14 detection lands on the exact date of the ALPHA3 trial futility announcement, confidence 0.9965. Not for real-time entries (BOCPD lags real change points by 5-20 observations); use for post-hoc regime labeling and to audit whether a pair's cointegration or a factor's edge broke partway through the lookback.

signal-decay Estimate the half-life of a candidate signal. Runs rolling 63-day IC vs 5-day forward returns over a 5-year window, fits an exponential decay to the IC series, and reports the half-life in trading days plus a full performance tearsheet (deflated Sharpe, Sortino, Calmar, ulcer, tail ratio, hit rate) on the signed-signal PnL. Motivated by 2024-25 factor decay research (Israel-Moskowitz-Ross, Falck-Rej-Thesmar 2024, Chen-Zimmermann factor zoo). Live on SPY mean-reversion: mean IC went from +0.06 early quarter to -0.16 recent quarter — the classic short-horizon mean-reversion effect has completely flipped sign this cycle.

rough-vol-forecast Rough-volatility-scaled vol forecast (Bayer-Friz-Gatheral 2016) across multiple horizons. Under rough vol, realized vol scales as h^H with H ~ 0.14 (Livieri et al. 2018 empirical) — much slower than sqrt(t). Reports the rough-vol forecast alongside traditional Brownian scaling and EWMA so you see the divergence directly. On SPY: 120-day vol drops from 11.66% (traditional) to 2.08% (rough) — a 5x reduction that changes how you size long-horizon scenarios.

Market context

prediction-market-monitor Kalshi prediction market monitor for Fed decisions, CPI, GDP, NFP, and other macro / market events. Motivated by 2025-26 post-election growth of prediction markets as leading indicators. Reports the implied probability at each strike, derives the cross-strike probability distribution when the event is a laddered strike-type (like KXFED-27APR-T4.25, T4.00, T3.75...), and computes the modal outcome and expected value. Live on Kalshi Fed decisions: October 2027 shows 63.5% probability of a 25bp hike; December 2027 shows a split market (37.5% hike vs 49% cut). Uses Kalshi's public read-only API; no Massive key required.

market-regime Daily macro briefing before you do anything else. SPY trend (above or below 20/50/200-day MAs, 5 buckets from uptrend_strong to downtrend_strong), VIX state with percentile rank vs trailing year, breadth proxy from the 11 sector ETFs (how many above 50-day MA), and 20-day RS leaders/laggards across sectors. Composite regime label (risk_on, risk_off, mixed_risk_on, mixed_risk_off, neutral) ships with explicit reasons so you see the evidence, not just the label. Run it at the open; it's the right anchor for everything else.

sector-rotation-signal market-regime reports current sector leadership as a snapshot. This one tells you how the leadership order has changed. Tracks 20- day RS rank for the 11 SPDR sector ETFs across a rotation window (default 30 days) and flags sectors that moved two or more positions up or down. Categorizes moves into growth / value-cyclical / defensive / rate-sensitive buckets and generates a plain-English theme read. Last week's run flagged risk-off: XLU up seven ranks and XLK down nine, defensives taking share from growth. Rank change is the leading signal; absolute standings are already priced.

macro-event-calendar Sibling to earnings-blackout. Forward calendar of the macro releases that reprice the whole book: FOMC decisions, CPI, PPI, NFP, ISM manufacturing and services, GDP, PCE, JOLTS, jobless claims, retail sales. Every event ships with its expected release date, ET release time, historical mean absolute one-day SPY move on that release type, and impact tier. Crowded days (two-plus events on the same date) get their own callout. FOMC dates are hardcoded from the official schedule; recurring events are pattern-derived and flagged as such so you can verify against the official calendar before sizing around a print.

historical-analog-finder market-regime tells you today's state. This one takes that state and finds K historical periods with the most similar setup, then reports the forward SPY return distribution across those analogs. Features are SPY-only (5/20/60/120-day return, above 50 and 200-day, RSI 14, realized vol, drawdown from 252-day high) with z-scored Euclidean distance. Deduplicates overlapping matches so one historical window can't dominate the K. The mean is not a forecast; the interquartile range is the honest read. Last run: 20 nearest analogs to the current tape gave a 75% hit-rate above zero at 60 and 252 days, median +14.3% at 252d, with mid-2024 and late-2021 pulling opposite directions in the K set.

single-name-vs-sector relative-strength ranks a name versus SPY, which conflates two things: is the name strong because its sector is strong, or is it pulling away from its own sector. This one splits them. Three RS legs across 5/20/60/120-day windows: name vs sector, sector vs benchmark, name vs benchmark. Classifies the name as leading, lagging, or diverging (name and sector pointing opposite ways). Surfaced when SOFI showed as a stable laggard at -4100bp vs SPY 120d while XLF was leading at +377bp 20d. The weakness was name-specific, not a financials problem. Divergence takes precedence over sign, so the tool catches the important case first.

commodity-cycle Is this commodity in a winning or losing macro setup, and which driver dominates. Rolling correlations against the dollar (UUP), the real-yield proxy (TIP minus IEF), and (for gold) miner divergence vs GDX and silver co-movement. Plus a momentum quintile ranked against the trailing year. Reads constructive, neutral, or headwind with the dominant variable named. On GLD 60d: headwind, dominant driver a strong dollar (60d corr -0.68) plus rising real yields, miners lagging -9%, momentum in the bottom quintile. Would have flagged the gold drawdown before price confirmed.

Trading and execution

options-flow You're scanning a watchlist for unusual options activity. Premium size, volume vs open interest, above-ask vs below-bid, repeat-strike clustering. Output is a tight stream of the top 10-20 prints with a sentiment tag per block. Yesterday's run surfaced a TSLA bullish read where someone sold $400 puts on the bid AND bought $385 calls above the ask. Two trades, same direction.

zero-dte-gamma Estimate net dealer gamma exposure (GEX) for same-day-expiry SPY / SPX / QQQ / IWM options and identify gamma pins. Motivated by 2024-25 research (Baltussen-Terhorst-Van Vliet 2024, Bhattacharya 2024) documenting that 0DTE options now drive systematic intraday moves through market-maker delta hedging pressure. Reports net dealer gamma, gamma regime (long/short), gamma flip strike, and top 5 pins with per-side call and put notional. Live on SPY 2026-07-13 expiry: $754.95 spot, dealers net short $1.25B gamma at the top pins concentrated at $759 call strikes (9,412 OI) — expect intraday moves to accelerate rather than mean-revert.

news-scanner You want today's notable news cross-referenced against the price reaction and sentiment. Each event ships with sentiment score, novelty score (is this a re-run or a new angle), and a price-vs- sentiment divergence flag. A "positive" article that the stock sold off on means the market already knew; that's a flag worth surfacing.

slippage-cost You hand the tool yesterday's executed fills. It pulls the microsecond NBBO at each trade time, computes slippage vs the inside, and flags fills that crossed the spread, printed off-NBBO, hit a wide spread moment, or showed adverse selection in the 30 seconds after fill. Measures fill vs NBBO at fill time, not arrival-price Implementation Shortfall. The exception report is short by design: only the broken stuff surfaces.

options-structure-analyzer You have a view but no structure yet. Tell the tool your thesis (direction bullish, direction bearish, vol long, vol short, or hedge), your horizon, and your target move. It enumerates the candidate options structures (long call or put, bull or bear spread, straddle, strangle, iron condor, protective put, collar), pulls the nearest expiry with priceable legs on both sides, computes each structure's payoff at your target, and ranks by payoff-to-capital. Not a recommendation. A structured comparison so you can pick the structure whose tradeoffs match your view, not just the one your platform happened to surface. On hedges, it reports the P&L improvement vs unhedged rather than a meaningless "percent of net premium" ratio.

hedge-suggester risk-report flags positions carrying most of the book's variance. options-flow shows what other traders are doing. Neither proposes what to do about a concentrated long. This one takes one long position and returns five priced option overlays (covered call, protective put, collar, put spread, ratio put spread) from the live chain, ranked by cost per dollar of downside protected. IV context from ATM implied vs 20-day realized volatility as a variance-risk- premium proxy, labelled expensive, fair, or cheap. Take fitted to the stated risk tolerance. On an ALLO 90-day run: collar priced as a net credit at -1.4% of notional, IV at 1.76x realized (expensive), tail-risk flag on the ratio structure. Not investment advice.

Risk and operations

portfolio-mark You need end-of-day marks for a position book. Run the tool. It pulls the snapshot per name, walks the fallback chain (last trade → snapshot last → minute close → day close → prior close), reports per-position confidence (high/medium/low), and flags any name where the mark looks stale or the spread is wide enough to need a manual check. Two modes: delayed REST for end-of-day reports, live WebSocket for intraday.

risk-report The book is already marked. Now what could happen to it? Run the tool and get VaR + Expected Shortfall at 95/99, max drawdown over the lookback window with peak/trough dates, the five worst historical days with per-name loss attribution, every position's share of the variance budget, and a Herfindahl-based concentration read. Pairs with portfolio-mark: marks tell you what the book is worth right now; this skill tells you what could happen to that value. Pass --mc for path-VaR: 10,000 correlated return paths simulated from the fitted covariance matrix, with normal or student-t innovations for fatter tails, over a configurable horizon. Emits cumulative return percentiles, path max-drawdown distribution, and P(loss > X%) at 5/10/20% thresholds.

mc-portfolio-simulator Standalone Monte Carlo forward P&L simulator. Give it a book and a horizon; get back the distribution of outcomes, tail scenarios, path max-drawdown, and probability of loss / gain at 5/10/20/30% thresholds. Uses the same covariance-matrix pipeline as position-sizer and risk-report --mc, so results are directly comparable across the three tools. Answers "given my proposed weights, what's the 5th percentile 60-day portfolio outcome?"

portfolio-macro-scenario risk-report and portfolio-review are descriptive of the past. This one is prescriptive under a scenario you name. Per-position OLS regression on TLT/UUP/USO/GLD daily returns, applies rates, dollar, oil, and gold shocks, aggregates to book P&L with a rough +/-1.64 sigma band. Ranks positions by absolute P&L contribution and factors by aggregate contribution. Rate shocks convert to a TLT return via an assumed effective duration of 17 years. On a 12-position book with +50bp / +2% dollar / +10% oil / -5% gold: book P&L -$17,130 (-3.9%), gold factor 40% of the loss, TLT factor 33%. Betas are historical and unstable, factors are collinear (rates, dollar, oil, gold co-move), and the CI band assumes residual independence, all noted in the caveats.

portfolio-rebalancer risk-report tells you which name is driving the risk. This skill tells you what to do about it. Feed it the same weights + book value; set a variance-share cap, weight cap, and churn cap; it returns a specific rebalance ticket list: which names to trim, which to add, exact dollar amounts, before-and-after weight, before-and- after variance share. Iterative solver with a proportional redistribution rule so one name's trim doesn't over-concentrate the next name. Refuses to churn more than the preset per rebalance so a single call cannot blow up a portfolio. Not tax-aware, not liquidity-aware, honest about both in the caveats. Turns "ALLO carries 66 percent of variance" into "sell $66k of ALLO, redistribute, portfolio vol drops from 21 to 15 percent."

corporate-actions-scanner Different job from corp-actions-reconciler. The reconciler checks whether your position file has the correct post-split share count. This one scans forward-looking: for a ticker or watchlist, it pulls SEC 8-K filings over a lookback window (default 180 days), filters to material items (offerings, private placements, splits, spin-offs, buybacks, M&A, restatements), cross-references Massive news for the headline, and computes the T+1 and T+5 price reactions. Built after a portfolio review missed an ALLO public offering (87.5 million shares at $2, 34 percent dilution) because news-scanner defaulted to a 24-hour window. The offering was 78 days old and dominating the current price, and no dedicated tool surfaced it.

corp-actions-reconciler An ops desk inherits a position file from 2024. Did the share counts get adjusted for AAPL's 4-for-1 split? GOOGL's 20-for-1? NVDA's 10- for-1? Run the tool. The exception report lists every position whose recorded share count doesn't match the expected post-split share count, with the source endpoint and verified-at timestamp on every flag.

t+1-settlement-prep You're an ops manager looking at tonight's trades. Which ones have settlement risk crossing this weekend? Which ones need a short-sale locate confirmed before tomorrow's cutoff? Which ones cross an ex- dividend date? The tool walks each trade against the US holiday and corporate-action calendar and flags six failure modes, with a suggested next action per flag.

Filings and ownership

8-k-scanner You have a watchlist. What material events hit those names this week? Run the tool. It pulls every 8-K disclosure across the basket, groups the tagged Items by filing, ranks by signal bucket (M&A / Strategic above Restatement above Material agreement above Regulatory above Leadership change above Capital above Earnings above Corporate housekeeping), and puts the highest-signal filings at the top with the supporting text quoted. Last live run caught RKLB's Iridium merger agreement, NVDA's Ajay Puri retirement + Nicholas Parker hire package, MSFT's Reid Hoffman board departure, ALLO's CEO succession, and NVDA's $11B notes issuance. One 45-day scan across 6 mega caps.

risk-factor-delta Did management add anything new to Item 1A this year? Run the tool. It diffs the last two 10-K risk-factor disclosures using Massive's pre-parsed three-tier taxonomy (primary/secondary/tertiary category per risk). Reports categories added, categories dropped, and categories where the supporting text materially changed (>= 25% length shift). Live run on AAPL's 2024 vs 2023 10-K surfaced 13 new risk categories concentrated in technology_and_information (AI risk, third-party data security, tech-vendor dependence), plus new trade/tariff and credit risk categories. Also caught 7 retained categories where AAPL trimmed language (pricing pressure -46%, IP -42%, ESG -39%): a defensive language shift the sentiment scorer independently confirmed.

filing-sentiment Score the Business and Risk Factors sections of a company's last two 10-Ks using the Loughran-McDonald finance dictionary. Reports rate per 10,000 tokens by category (negative, uncertain, litigious, constraining, modal-weak, modal-strong) prior vs current with a shift label. Answers "did management's language get more defensive?" Live run on AAPL showed negative language up 33% in the Business section and modal-strong ("must", "will", "always") language down 38% in Risk Factors: management getting less assertive and more hedged. Pair with risk-factor-delta for the structural view; this handles the tone.

insider-flow Are insiders buying or selling this name? Run the tool. It pulls every Form 4 for the ticker over the lookback window, classifies each transaction (conviction buy P, discretionary sale S, scheduled 10b5-1 sale, routine comp A/M/F, non-informative), filters 10b5-1 out of sentiment, detects cluster buys (>= 2 insiders in a 14-day window worth >= $100k), and emits a bullish/ bearish label backed by net conviction dollar flow. --exclude- directors cuts out VC/PE board reps unwinding fund positions (which structurally look bearish but carry no operator signal). Live run on RKLB showed a $27M director sale from a VC rep that, once excluded, still left the general counsel and president as net sellers.

manager-portfolio-diff What did Buffett/Klarman/Burry do last quarter? Run the tool with an alias (--filer berkshire) or a CIK. It pulls the two most recent 13-F snapshots for the fund, aggregates by CUSIP across joint filings, and reports initiations, exits, adds (>= 25% share change), trims (<= -25%), and portfolio value delta. Live run on Berkshire caught Buffett exiting Visa, Mastercard, UNH, Dominos in Q1 2026 while adding Alphabet +204% and initiating Delta Air Lines at $2.65B. Alias set: berkshire, baupost, renaissance, bridgewater, third-point, pershing, tiger-global, scion, appaloosa.

guidance-tracker Has management been raising or cutting guidance this cycle? Run the tool. Uses Benzinga Corporate Guidance to compare each event's midpoint against the prior figure for the same fiscal period, labels as raised / lowered / reaffirmed / initiation, groups by fiscal period, and reports trajectory. Requires the Benzinga Corporate Guidance add-on (approx $99/month); the skill emits a clean NOT_AUTHORIZED caveat when the entitlement is missing.

analyst-tracker Track sell-side positioning on a name. Uses Benzinga analyst ratings to pull every rating event over the lookback window, classifies each as upgrade / downgrade / initiation / reiteration / PT-change / drop-coverage, aggregates the latest per firm, and reports the consensus median price target plus buy/hold/sell distribution. This is the sell-side lens pair with insider-flow (internal vs external sentiment on the same name). Live NVDA run over 180 days: 26 firms all Buy, consensus PT $308.50, big