API Console Documentation

One-call company snapshot (Console DES): identity and SIC-based sector; last price, 52-week range and 1D/1M/3M/YTD/1Y total returns; guarded market cap, shares outstanding, P/E, EV and dividend yield; 13F holders and institutional % of shares outstanding; next report date (confirmed vs projected); TTM revenue, operating income, net income, EPS and free cash flow from de-cumulated quarters; and flags (activist 13D, 13F cluster entry/exit, accumulation streaks, credit stress, going-concern doubt, open exchange-listing deficiency, short squeeze/spike); FINRA short interest (% of shares out, days to cover). Percent fields are percents (12.5 = 12.5%).

Event calendar (Console EVT). Upcoming: next earnings (vendor-confirmed or projected with window and confidence), 10-K/10-Q filing deadlines from fiscal year end + filer status (approximate, late flagged), expected proxy, declared dividends (amount, change, ex/record/pay dates) or else an estimated next ex-dividend, catalysts. Past: earnings with surprise and day-1 move, split-adjusted dividends and splits, filings with 8-K item labels, auditor changes, Item 1.03 8-Ks and credit-stress signals. Predecessor registrants are included (e.g. DINO's HollyFrontier history). Estimates carry is_estimate=true.

Open a filing from our store by accession number: form, company and CIK, filed / period dates, the list of documents in the submission (sequence, type, filename, description) and its item sections (10-K, 10-Q, 8-K, 20-F, 40-F). Returns the text (or HTML) of the main document, a chosen exhibit (seq) or one item (section), in chunks: content_length, offset and next_offset page through long documents. Never returns SEC.gov links.

Catalyst Calendar (Console app): market-wide agenda of dated events that move stocks — FDA decisions (PDUFA) and advisory committees, clinical readouts, device approvals and panels, CMS coverage and rate notices, tariffs, foreign permits, IPO lock-ups, debt maturities (off by default), exchange-compliance deadlines still open, and expected earnings dates (projected; by default $10B+ only, since ~2,400 a month bury everything else).

Filings: every filing the company made, newest first — all forms, including Forms 3 / 4 / 5 filed about its insiders — each with its exhibits.

Red Flags (Console RF): accounting, listing, risk and related-party trouble for one company.

The sentence classifiers that score every SEC filing sentence: name, readable label, what it detects, source (human-written or auto-discovered), direction (positive / negative where it has one), version and whether it is in the licensed signals feed; the curated ones carry their theme. Use a name with get_sentences or screen_companies. Without a keyword or source, lists them all.

The sentences that made a classifier fire, with the filing they came from: the sentence, its score (0–1), classifier, form, filing date and accession — for a company (with one or more classifiers) or for one filing. Best score first (ties by sentence), so quote these as the evidence for a signal. Scores rank over the whole history unless you set since.

Filing signals: every filing the company made or was the subject of (13D/G about it) over 3 years, including predecessor registrants, newest first, each with the curated classifier tags found in it (theme, label, direction, score) and the best evidence sentence for each. Also returns theme totals (last 12 months vs the 12 before), counts per form group and the company's latest filing signal profile. Structured-note prospectuses (424B / FWP) are capped at 100.

Screen every SEC filer on the signal profile of its latest filing: composite, distress and positive signal scores, how much the profile changed (mutation) and how unusual it is (entropy surprise), the top rising classifiers, insider buying, bond-spread widening and 8-K activity in the last 90 days (items, counts and the recent events with their accessions). One row per company with its CIK, ticker, SIC sector and the accession of the filing scored. Scores grow with company size and filing length; compare like with like.

Search SEC filing sections by meaning, not keywords: the passages closest to what you describe, across every form in the corpus (10-K, 10-Q, 8-K, proxies, prospectuses), optionally for one company, some forms or sections, and a date window. Each result is the filing (accession, form, date, company), the section (item and heading), a snippet and a relevance score (higher is closer; not on a 0–1 scale). Without a company the search covers the last 3 years unless you set since. Open the full section with get_filing.

Which companies rewrote their risk factors (Item 1A) the most: for each company's latest annual or quarterly report in the year, the risk-factor paragraphs that are new against its prior year's filing — how many, how many rated high or medium materiality, and their risk categories — with the filing's accession and the prior filing's date. Covers risk factors only (Item 1A), not MD&A or legal proceedings. The per-company paragraphs are in company_red_flags.

IPOs priced in a window, newest first, from the final prospectus (424B4): company, ticker, exchange, offer price and the filed range, gross proceeds, implied market cap, shares offered, dual-class flag, lead underwriters, SIC sector and the prospectus's accession. SPACs are left out unless you ask for them; de-SPAC listings are flagged. Coverage starts in 2020. Deal detail: sections_ipo_offering.

One company's IPO deal card from its final prospectus: offer price and filed range, shares offered (primary, secondary, over-allotment), gross and net proceeds, implied market cap, exchange and ticker, share classes and voting ratio, every underwriter with its role, use of proceeds, and the lock-up — days, expiry, share of float locked and each lock-up tier (who is covered, early-release terms). Coverage: IPOs from 2020.

The risk factors of a company's IPO prospectus (the final 424B4, else its first S-1 / F-1), in the order the prospectus lists them: each risk factor's heading, the group it sits under (e.g. 'Risks related to our business'), and the opening of its text where the heading is found in the section.

Ownership Summary (Console OWN) for one 13F quarter: institutional % of shares outstanding and of float, insider % (Form 4), float, top-10 concentration, holder count and quarter-over-quarter change (new / exited / increased / decreased, computed from our own prior-quarter comparison), active vs passive %, investor-type mix, holder quality (shares held by managers beating vs lagging the S&P 500 over 2 years), a 9-quarter history and the top 20 holders. Percent fields are percents. Insider data is current-state, so historical quarters are 13F-only.

Combined holder register (Console HDS): 13F institutions (canonical manager, share classes merged when share_classes=all) plus Schedule 13D/G beneficial owners and Form 4 insiders, one row per owner, largest first, with % of shares outstanding, market value, quarter-over-quarter change, investor type, passive flag and the filings behind each row. Historical quarters are 13F-only. Returns up to limit rows; n_13f_holders is the full 13F holder count.

Ownership Activity (Console OA): 13F buyers, sellers, new positions and exits in one quarter (top 25 each), computed from our own prior-quarter comparison (never the upstream 'new' flag), with net shares, passive vs active net, holder-quality net flow, and the notable-manager flow trend over 8 quarters.

FINRA short interest (Console SI): latest biweekly settlement with short shares, % of shares outstanding and of float, days to cover, change vs prior settlement and year ago, 1-year high/low, squeeze / spike / covering flags, and the settlement history restated for stock splits. Ticker reuse (a symbol that belonged to another issuer) is cut out of the history.

Insider Liquidity (Console LIQ): when insiders can sell and have said they will.

Insider buy track record (Console INS buyers' records · person Buy Track Record): how open-market buys did afterwards against the S&P 500.

Insider Track Record app (Console 'Buyers'): market-wide, recent open-market buys by insiders whose past buys beat the S&P 500.

Corporate Activism (Console ACTV): every 13D/13G holder's latest position (form, % of class, shares, activist flag, campaign status, extracted intent and tone), the full 13D/G filing trail with intent, and campaign events from 8-K / proxy filings (proxy solicitations, board demands, cooperation agreements, contested elections, poison pills, withdrawals) with outcome and accession.

Proxy Votes (Console PV): shareholder-meeting results for 10 years and what's on the next ballot.

Bond yields (Console CAPS / DEBT): each fund-held bond of the issuer (and its predecessors) — CUSIP, coupon, maturity, fund-held par and value, number of funds, the month-end CONSENSUS price across funds' N-PORT marks (not a trade or dealer quote), its dispersion, yield to maturity (semi-annual, fractional periods), 1-month and 3-month price change and a distressed flag (YTM >= 12% or price < 50) — plus par-weighted coupon / yield / price and the next maturity. Uses the newest settled month; report_date says which. Matured bonds are dropped.

Private Capital (Console PVT): one company's private financing — Form D offerings with amendments grouped (offered, sold, investors, exemptions, filings by accession), funding rounds deduplicated across outlets ('duplicates' lists merged copies) with investors and leads, valuations only when not flagged aspirational or outlier, Reg CF offerings, related persons, placement agents (CRD), totals that never blend equity with debt, and the days a Form D preceded the news. Pass ticker / cik / cusip / q, or company_id for a private company.

Private funds (Console ADV, Form ADV Schedule D 7.B(1)) of a registered investment adviser: each fund's type, gross asset value, minimum investment, beneficial owners, % non-US, 3(c)(1)/3(c)(7) exclusion, master/feeder structure, audit, domicile and its auditors / prime brokers / custodians / administrators / marketers; fund counts and GAV by type; and a provider league table across the adviser's current funds. Pass crd, or a 13F manager cik / name — a firm with several advisers (BlackRock, Goldman) lists them in advisers and opens on the one with the most private funds; pick another with crd.

Capital Graph of a company (Console Capital Graph): who is linked to it, from the filings, as one network — Owners (13F holders at the latest quarter, 13D and 13G filers, N-PORT fund holders over the last ~4 months), People (board members, officers, Form 4 insiders over 3 years, founders), Corporate (acquisitions both ways, subsidiaries, parent), Capital (private-round investors, Form D placement agents) and Commercial (counterparties to filed material contracts, both ways).

Capital Graph of a 13F manager (Console holder Capital Graph): Public equity (13F positions at the latest quarter, 13D and 13G stakes), Private capital (private-round investments, Form D placements) and Funds (the N-PORT funds it manages, by net assets).

One node of the Capital Graph by id — the next hop from an item of company_graph, holder_graph or another graph_node: a person (board seats, officer roles, Form 4 insider roles, companies founded, career, education, connections), an investor (as holder_graph), a fund (holdings at the latest N-PORT month, adviser), an institution (alumni) or a company (as company_graph).

Geographic revenue (Console GEO): revenue by country and region exactly as the company tagged it in XBRL — never estimated; a filer's 'rest of world' stays 'rest of world (as reported)' — turned into non-overlapping slices (value, % of revenue, year-over-year change, multi-year trend), with super-region (Americas / Europe / Asia-Pacific / Africa-Middle East) and developed / emerging / frontier roll-ups, the tag and accession used, and coverage: the share of revenue the disclosure explains plus xbrl's own coverage check (coverage_pct, is_anomalous). Fiscal-year revenue comes from de-cumulated quarters. History starts ~2020 (inline XBRL).

Supply chain (Console SPLC): the company's own 10-K / 10-Q concentration disclosures (named and unnamed customers and suppliers with % of revenue / purchases / receivables, each tied to its filing), the companies that name this one as a customer or supplier in their filings (e.g. Cirrus Logic: Apple = 90% of net sales), and news-sourced relationships (customers, suppliers, partners, distributors, licensors, JVs, competitors) with mention counts. Filing data leads; news relationships are noisier (direction errors exist upstream).

Material contracts (Console MC): every agreement the company (and its predecessors) filed as an exhibit since 2016 — title, type and family, execution / filing date, counterparties (companies and people, with roles), amount, term, amendment flag, a key quote, and the clause types it contains (change of control, non-compete, indemnification…) — plus facets: agreements per family and year, clause types with counts, and named people (employment / severance terms). Each agreement opens with get_filing(accession, seq). Clause passages are located only for merger agreements so far; other types list the clause type.

Capital structure (Console CAPS): price, cover-page shares (with accession) and the guarded market cap; debt (short / long term), cash, leases, equity and EV; TTM revenue, EBIT, EBITDA (only where D&A is tagged), net income and interest; leverage and coverage ratios; 8 quarters of history; the filer's own debt schedule from its latest 10-K / 10-Q; debt issued per 8-K Item 2.03 (issuances, not balances); fund-held bonds and the maturity ladder (N-PORT, fund-held par only); credit-stress signals. Banks and insurers (SIC 60-64) are flagged financial: EV and debt-based ratios don't apply.

Comps (Console COMP): trailing valuation and operating metrics for the company and its peers — market cap (guarded), EV, P/E, EV/Sales, EV/EBIT, EV/EBITDA, P/B, revenue growth, gross / operating / net margin, ROE, dividend yield, 1-year total return, net debt / EBIT — with peer medians that exclude the company. Default peers: up to 12 same-industry (SIC) majors by market cap; or pass your own. Trailing only (no estimates); EV multiples hidden for banks and insurers.

Dilution & Capital Markets (Console ECM): how a company raises money and what it does to the share count, 5 years.

ECM Monitor (Console app): market-wide equity capital markets and dilution, listed companies only.

Sector monitor (Console Sector Intelligence): the 11 SIC-based sectors side by side over the screening universe (listed issuers with 10+ 13F holders, funds excluded): companies and total market cap, 1-day / 1-month / 3-month / YTD / 1-year returns, 1-month breadth (% of companies up), notable-manager 13F net flow, cluster entries / exits, median change in institutional % held and % of companies gaining holders, insider buying and selling (90 days), median short interest and squeeze setups, 13D filers and activism campaign events, companies reporting in the next 14 days, and median valuation — plus, for the window, the insider sector pulse (discretionary open-market trades by SIC) and new Form D raises mapped to sectors.

Sector detail (Console Sector Intelligence): one sector — its 4-digit SIC industries with the same aggregates as sector_monitor (biggest first), Form D raises (window total, by Form D industry, the 15 largest offerings and 12 months of monthly totals), the insider pulse (window total and the 12 industries with the largest net flow), 1-month movers up and down ($1B+), notable-manager 13F buys and sells, cluster entries / exits, the largest insider buying (90 days), the most-shorted names ($300M+), recent 13D / activism, earnings in the next 30 days, catalysts in the next 90 days (FDA dates, lock-ups, …; by accession), the cheapest and richest trailing P/E ($1B+), and the constituents by market cap — paged with limit / offset (total counts them). Aggregates: returns (chg_1d, r1m, r3m, ytd, r1y, in %) are market-cap weighted; valuation figures (*_med: pe, ev_sales, ev_ebit, ev_ebitda, pb, rev_growth, gross / op / net margin, roe, div_yield) are medians of trailing-twelve-month values; flows are sums.

Industry detail (Console Sector Intelligence): one 4-digit SIC industry — its aggregates, the insider pulse for that SIC, movers, 13F flows, insider buying, shorts, activism, upcoming earnings and catalysts, cheap / rich trailing P/E, constituents (paged with limit / offset), plus its sibling industries in the same sector (companies and market cap) and the sector-wide aggregate for comparison. Aggregates: returns (chg_1d, r1m, r3m, ytd, r1y, in %) are market-cap weighted; valuation figures (*_med: pe, ev_sales, ev_ebit, ev_ebitda, pb, rev_growth, gross / op / net margin, roe, div_yield) are medians of trailing-twelve-month values; flows are sums.

Price history (Console Charts): daily bars from atlas market.prices_v — o / h / l / c split-adjusted to today's shares, a total return (adj_close: splits + dividends reinvested), raw as traded that day, v volume.

Chart events (Console Charts): the filing-native markers Console draws on a company's price chart, each with the accession it came from — earnings (beat / miss vs the company's own trend, day-1 move), cash dividends (split-adjusted to today's shares) and splits, insider open-market buys / discretionary sells / 10b5-1 plan sales (one marker per day and kind), initial 13D / 13G and activists' 13D amendments, activism campaign events, new Form D offerings, debt issued (8-K Item 2.03), material commercial / M&A / capital-markets / real-estate / credit agreements, auditor changes, 8-K Item 1.03 bankruptcy, and US macro release days (CPI, jobs, GDP, PCE, FOMC). Each marker: d (date), t (type), label, detail, acc, value, tone (+1 good / -1 bad). Predecessor registrants are included (e.g. HollyFrontier for DINO). counts is per type over the whole range.

People & Board (Console MGMT): current executives (ranked by title) and directors, rebuilt from Form 4 filings, the latest proxy and 8-K 5.02 appointments/departures — with title, age, latest-year pay (proxy), Form 4 holdings (direct / indirect / derivative), other board seats and the source filings; plus recent appointment / departure events and, optionally, former officers.

Person profile (Console People): one person, every section always present (empty when MARS has nothing yet).

M&A history (Console M&A History): one row per real deal the company was in — as acquirer, target or seller — SEC first: filed merger / stock-purchase / asset-purchase agreements (2016+, including agreements filed by the counterparty), then news, de-duplicated (the same deal reported many times is one row; investments mis-typed as acquisitions are left out).

Resolve a gene / protein / modality / indication alias to its canonical name (HGNC symbol for genes), with every candidate: PD-1 is both PDCD1 and SNCA. Exact match on the normalized alias first (resolution exact_normalize); when that misses, a substring match on names and aliases (substring). Pass a canonical gene to bio_target_base_rate, bio_target_trajectory or bio_landscape.

A company's drug pipeline from its filings, point in time.

Who is working on a target or an indication: every company's assets on that gene (or matching that indication) in its latest pipeline snapshot on or before as_of_date — company (CIK, ticker, name; name_resolved false for private or delisted sponsors), asset, modality, highest phase, snapshot date, accession and snippet. Also n_companies and n_assets over the whole result. Resolve aliases with bio_entity_resolve first.

One page before an FDA decision: the company's next PDUFA or advisory-committee date in the next year (from the catalyst calendar), and a one-line headline per context pillar, each with a drill link to the detailed tool — prior CRLs (count, most recent reason, drug and date), the post-CRL base rate for that reason (median 252-day abnormal return, share down, approved-later rate), the base rate for the lead program's target gene, management (officers with a prior approval elsewhere), manufacturing risk (OAI inspections, latest classification) and the pipeline (assets in the latest snapshot). A pillar that fails says available: false; the rest still answer.

Clinical-trial readouts the company disclosed in its filings, newest first: drug, phase, outcome (MET / MISSED / MIXED), p-value, endpoint, the sentence it was read from, filing date and accession. pvalue_outcome_conflict flags rows whose p-value contradicts the outcome.

FDA drug and biologic approvals (CDER and CBER), newest first, for one company or the whole market: application and submission number, action date, sponsor, brand, active ingredient, modality, source, biosimilar flag, and the matched public company (CIK, ticker, match method and confidence).

What happened after FDA Complete Response Letters: by_reason — per reason category (CMC / manufacturing, efficacy, safety, bioequivalence, clinical design, labeling, inspection, other) the number of CRLs, median and mean 252-day and median 126-day SPY-adjusted abnormal return, share down at 252 days, and the approved-later rate among CRLs with a known resolution, computed over every matching CRL; events — each CRL (drug, company, reason, date and its precision, abnormal returns at 21 / 63 / 126 / 252 days, eventual approval), newest first, paged.

A sponsor's prior FDA Complete Response Letters known before as_of, newest first: application, letter date, reason category with its confidence, the key deficiency sentence, the disclosing filing's accession (for CRLs disclosed in filings), whether it was approved later (only when that approval also predates as_of), and flags for the same application or the same molecule in another formulation. An empty answer is not evidence of no prior CRL (coverage_caveat).

FDA warning letters, newest first: company on the letter, issue and posting dates, issuing office, subject, whether there is a response and a close-out letter, and the matched public company (CIK, ticker, match method and confidence). For inspection results see bio_inspections.

FDA inspections and their classification, newest first: site (FEI number, legal name, city, state, country), fiscal year, project area, product type, classification (OAI / VAI / NAI) and date, whether citations were posted, whether the site is a contract manufacturer, and the matched public company.

The historical success rate of drug programs on a target gene up to as_of_date: programs resolved, positive (approved, or acquired while active), approved, acquired, discontinued, and the success rate; verdict is insufficient_priors under 3 resolved programs. With exclude_cik the rate is recomputed without that company. methodology_note says exactly what counts as success.

How crowded a target is over time, quarter by quarter: companies and assets on the gene, the furthest phase reached (and its rank) and the leading companies (CIK, ticker, phase).

A biotech's de-risking score for the latest quarter on or before as_of_date, with its components (assets in the pipeline, an approved drug, market cap, cash runway in years, partnership, target crowdedness) and their sources. A sizing / volatility signal, not a direction call: read caveat.

Cash runway quarter by quarter, newest first: years of runway (cash / annual operating burn), cash, quarterly burn, free cash flow, quarter-over-quarter change and the declining streak; summary flags a deteriorating runway (3+ quarters declining).

When a company's drugs lose exclusivity, soonest first: brand, ingredient, modality, loss-of-exclusivity date and its source (Orange Book patents, or the 12-year biologic rule — loe_source_is_real_patent_data), latest and earliest patent expiry, patent count, substance patent, latest exclusivity expiry, Orange Book edition and application number.

A biotech's insiders (Forms 3 / 4) with their record across every company they have been an insider of: role here, officer / director, first and last filing here, and track_record — companies, top role, and how many reached an approval, were acquired, failed or are still active. Officers first, then by approvals.

One person's biotech track record: every company they were an insider of (role, first and last filing, the company's outcome: approval / acquired / failed / active) and the counts. Find them by name or CIK; most approvals first.

The power universe: power-relevant SEC filers (utilities and their operating companies, independent power producers, LNG, midstream and pipeline filers), survivorship-complete with successor / predecessor links. With a ticker, CIK or name: that one entity's card (role and its confidence, ISO, generating capacity and mix, share regulated, parent, HQ state, status, first and last filing). Without one: the list, largest generating capacity first, filtered by role, ISO, activity and SEC filer status.

One utility's power panel from its filings: its identity (role, ISO, share regulated, active), the last 10 years with data — capacity added and retired (MW, by fuel), net change, firm forward demand and its year-over-year change, outage count and MW affected — its rate cases (commission, case, decision date, allowed ROE with a suspect flag outside 3–20%, revenue increase, outcome, accession), and coverage: which datasets have any rows for it, so an empty panel isn't read as a finding. The company is resolved in the power universe (utilities and their operating companies, IPPs, LNG and midstream filers) by ticker, CIK or name.

Generating units being added or retired, as disclosed in filings, soonest target date first: filer (CIK, name, ticker), unit, addition / retirement, fuel, MW, target date, status, state, ISO, disclosed capex, accession and snippet — the latest filing per unit and event (with as_of, as known on that date). Market-wide, or one filer. For the aggregate curve use sec_power_supply_curve.

Forward large-load demand utilities disclose (data centers, industrial, crypto, hydrogen, EV charging): each load's filer, counterparty, expected MW and GWh, start year, status, geography, ISO, accession and snippet; and summary over everything that matches (not just this page): firm vs pipeline MW and counts, firm loads with no MW, and the split by start year. Firm = contract, in service, executed, firm or permitting.

The net forward supply curve across all filers: per fuel and target year, MW added and retired, the number of units each way, net MW, and a representative accession and snippet. Point in time with as_of (plans filed later don't leak in).

Plant outages material enough to be disclosed in a 10-K / 10-Q / 8-K, newest first: filer, unit, fuel, MW affected, outage type, start and expected return, cause, accession and snippet (latest filing per unit, outage type and start date).

Utility rate cases from filings, latest ruling first: filer, utility, commission, case, ruling and effective dates, revenue increase, allowed ROE, outcome, summary, accession and snippet (latest filing per case); and roe_distribution_by_commission over every matching case (not just this page): cases, mean, median, min and max allowed ROE in percent.