PineGap MCP Documentation

Extracts the performance drivers and explicit weightings that determine executive incentive pay — not the dollar amounts — from the latest proxy filing (falling back to 20-F/Annual Report). Produces a concise markdown summary separating short-term vs long-term incentive drivers for named executive officers, grounded solely in filing text, and can include the prior fiscal year for comparison.

Retrieves the latest primer sections for a company — a deep-dive structured overview covering business model, segments, competitive dynamics, and financial framework. By default it returns all available sections; pass a list of section types via sections to scope the response for token-bounded retrieval across multiple calls. The COMP_SHEET section is the largest and is only built when requested explicitly.

Scores earnings quality risk across six accounting dimensions using the most recent 10-K or 10-Q (falling back to 20-F/Annual Report), returning a per-flag status and filing-specific evidence for each raised risk. The six dimensions are revenue-recognition policy changes, disproportionate contract-asset growth, rising DSOs / slower collections, changes in significant accounting estimates, aggressive cost capitalization, and non-GAAP creep. Each flag carries a prior-period comparison to distinguish new, persistent, and resolved risks.

Flags financial and accounting risks from the latest 10-K/10-Q — including liquidity stress, earnings-quality issues, aggressive accounting or non-GAAP adjustments, control weaknesses, covenant pressure, and balance-sheet impairment triggers.

Surfaces corporate governance and shareholder-rights red flags from proxy disclosures (DEF-14A) — board independence/oversight issues, problematic pay practices, related-party transactions, weak say-on-pay signals, and anti-shareholder provisions.

Generates a portfolio-style attribution paragraph explaining why a stock outperformed or underperformed over a selected period, incorporating earnings transcript context and quantitative return data.

Retrieves a company's comp sheet tables — competitor lists, metric columns, periods, and resolved actuals/estimates cells. Supports three modes: no-cells (tables and competitors only, smallest payload — use first to discover the available periods), full (tables plus all cells), and by-period (tables plus cells for one required period). Because a full dump can be large, call no-cells to list periods, then by-period per period to reconstruct the same data in token-bounded chunks across calls.

Retrieves company-specific financial metrics and operational KPIs for a single stock from a natural-language query. AI selects the most relevant metrics from the company's actual available parameters, then fetches actuals, consensus estimates, earnings surprises, and estimate revision changes. All monetary values are in USD.

Fetches key company profile fields for a ticker — market cap, currency, 52-week range, IPO date, geography, beta, and average volume from the market data provider.

Retrieves conference and investor-day-style document summaries for the inter-quarter window following a specified earnings period. The window starts on that period's earnings call release date and ends at the next earnings call. Pass generate:true to enqueue summaries for documents in the window that don't have one yet.

Retrieves short AI-generated takeaways (one paragraph per conference) for investor/analyst conferences that occurred after a specified earnings call. The window opens on the earnings call release date and has no upper bound, so older periods may bleed across multiple subsequent quarters.

Fetches dividend history for a ticker — ex-dividend dates, payment dates, and dividend amounts.

Returns the historical earnings-day stock-price reaction table — 1-day and 5-day post-earnings price moves for recent fiscal periods, ordered most-recent-first. Reaction and surprise values are null when no genuine data exists for a period; nothing is fabricated. found is true only when the ticker resolved and at least one reaction is available.

Retrieves a comprehensive post-earnings analysis for a completed earnings period. Analyzes earnings call transcripts and press releases to produce a narrative summary of key themes, a metrics table comparing actuals vs consensus (revenue, EPS, and key metrics), and guidance highlights. Pass generate:true to enqueue recap generation when none is stored yet.

Returns a structured post-earnings scorecard — price reactions (1-day, 5-day), implied move, beat/miss metrics, estimate revisions, and guidance direction — for one or more tickers at once. Lighter than earnings-recap; pass metricsOnly:true to skip the AI key-debates/guidance-commentary and return only the numeric cells.

Retrieves consensus estimate history for a ticker across one or more Pinegap Data IDs (pgids) and calendar years, comparing the consensus as of a historical date against the latest consensus — in a single call. Replaces repeated per-metric/per-year revision-chart calls. A year with no revision activity in its lookback window returns null for that side with a row-level note rather than failing the whole request.

Retrieves a company document's full text — earnings call transcripts, press releases, 10-K/10-Q, 8-K, and SEC filings (Form 3/4/5/13F). Call either by documentIds (exact stored documents, re-openable from citation IDs) or by ticker + documentTypes with optional periods. A single document returns paginatable content; several return a documents[] array. deliveryMode defaults to content; viewer and download return a URL only (no text).

Advanced guidance extraction that captures all forward-looking statements from management across earnings calls — from precise quantitative targets (revenue, EPS, margins) to qualitative directional commentary — for one or more tickers at once. Classifies each item as RAISED, LOWERED, MAINTAINED, or NEW. Pass generate:true to enqueue extraction when none is stored.

Retrieves the most discussed topics and key debates from earnings calls, for one or more tickers at once — the themes management and analysts focused on, surfaced through frequency analysis and importance weighting. Each debate summary embeds inline [n] citation markers linking to the source text.

Returns fiscal period information for a ticker. In latest mode (no year supplied) it returns the most recent period for which content is available. In exists mode (year + quarter/half supplied) it confirms whether a document exists for that specific period. Defaults to earnings calls; pass documentType for another type.

Fetches the latest market quote for a ticker — current price, daily change percentage, session high/low, and 52-week range.

Unified access to investment-focused podcasts — browse podcasts and episodes, fetch transcripts (JSON or text), retrieve AI-extracted equity insights, and scan one or more tickers for mentions across episodes. The mode field selects the operation.

Returns standardized quantitative metrics for one or more tickers from a natural-language query — historical actuals and forward estimates, valuation multiples (P/E, EV/EBITDA, Price/Book, FCF yield), market data (market cap, enterprise value), consensus estimates with dispersion, estimate revisions, and earnings surprises. All monetary values are in USD.

Returns quotes from other companies' recent earnings calls and press releases that mention or link to a target ticker. Each read-through includes inferred impact direction and relationship context — useful for building a mosaic across the value chain.

Returns risk factors from a company's 10-K/10-Q filing. Mode 'new' returns risks classified as NEW or UPDATED in the specified filing. Mode 'active' returns the full active risk set through the resolved filing.

Retrieves the latest running screener by default, or lists all screeners for the user, returning progress, final tickers, metrics, and optional per-company report content. When a screener has completed, pass watchlistName to save its results into a new watchlist.

Fetches short-interest history (% of float, bi-monthly settlement data) for a single ticker over the trailing ~89 days. Bundles a renderHints block for drawing a step-after time-series chart. found is true only when the ticker resolved and the window has at least one point.

Retrieves stock information by ticker symbol, Bloomberg ticker, Nasdaq ticker, or company name. Returns company metadata including ID, name, exchange, country, sector/industry classification, reporting frequency, and external data-provider identifiers. Ticker lookup tries an exact match then suffixed variants (e.g. MND_US); name search is fuzzy. Use it first to verify a ticker before calling other tools.

Retrieves the ordered value-chain peer list for a ticker — the related companies (customers, suppliers, competitors, partners) as configured in pinegap for the authenticated user.

Analyzes capital allocation strategy covering liquidity, reinvestment, debt philosophy, shareholder returns (dividends/buybacks), and M&A approach. Grounded in the latest 10-K and proxy filing.

Generates a competitor identification and competitive landscape analysis from the latest 10-K/20-F filings. Covers market position, key differentiators, and competitive dynamics.

Generates a comprehensive pre-earnings analysis for an upcoming report. Combines consensus estimates and historical beat/miss patterns, recent news summarized by AI, and an AI-generated preview takeaway. Best used 1–2 weeks before an earnings call.

Generates detailed profiles for the C-Suite leadership team — synthesizing proxy statements, 8-K filings, and web research for background, tenure, and prior roles.

Fetches management follow-up questions for the latest earnings call, grouped by thematic section (financial performance, operations, strategy, etc.). Useful for preparing for management meetings or building a monitoring framework.

Pulls verbatim management quotes from a single source document — the latest earnings call transcript, or the latest press release as a fallback — organised into 12 operational categories such as volume, pricing, retention, sales efficiency, capital allocation, and forward indicators. Every bullet is a direct excerpt; nothing is paraphrased or generated.

Generates a markdown proxy review covering board oversight and composition, ownership and voting control, executive compensation structure and performance linkage, and notable shareholder proposals. Grounded solely in DEF-14A filing text.

Summarizes the voting matters in a company's latest DEF 14A proxy statement — each proposal and the board's own stated position on it — for research. This is descriptive: it does not recommend how to vote and returns no AI vote recommendation.

Creates or continues an AI-powered screener conversation. Describe the investment criteria in natural language; the screener interprets them, may request follow-up answers via elicitation (or return questions to ask the user), and runs asynchronously. Poll screener-status with the returned conversationId for results.

Tracks how management sentiment around specific keywords shifts across consecutive earnings calls, for one or more tickers at once. Compares the current and previous periods to surface changes in tone, emphasis, confidence, and language, with verbatim supporting excerpts. Omit keywords to return each ticker's stored curated keyword set.

Detects unusual, novel, escalated, or suppressed management disclosures by comparing the latest earnings call or press release against prior periods, for one or more tickers at once. An AI pass surfaces disclosures that are newly added, escalated, withdrawn, or downgraded.

Verifies the MCP server is running and its database connection is active. Use to diagnose connectivity issues before making other requests.

Creates a new watchlist for the current user with an optional initial set of tickers. When enableNotifications is true, all supported stock-level notifications are turned on for each added ticker.

Returns all watchlists for the current user. Optionally supply documentTypes to enrich each ticker with the latest release date for the selected document types.

Updates an existing watchlist's name and/or ticker list for the current user. Supplying tickers replaces the entire existing ticker list.

Deletes one of the current user's watchlists, removing it and its stocks from all views. The default watchlist and agent-managed watchlists cannot be deleted, and a watchlist that does not exist or belongs to another user returns a not-found error.