Open-source, self-hosted market research terminal: SEC filings, market data on your own vendor keys, and 37 read-only MCP tools for your own AI agent.
AlphaDesk exposes 26 financial data tools with consistent naming (verb_noun pattern: market_tape, price_chart, find_symbol) and detailed descriptions (avg ~180 chars, well above 10-char minimum). All tools are READ_ONLY, reducing security risk. However, critical gaps emerge: input schemas are visible for only ~12 tools in the provided source; the remaining 14 tools lack explicit schema definitions in the code excerpts provided. Parameter descriptions are present but inconsistent, some tools (price_chart, movers) have rich constraint documentation (e.g., 'max 400', 'default 20'), while others (my_board, sector_performance) have empty input objects with no guidance. Output schemas are entirely undocumented, no tool declares what fields it returns, forcing LLMs to guess. Error handling is absent: no recovery guidance, no categorization of retryable vs. fatal errors. Composition is strong: tools chain naturally (find_symbol → quote → financial_statements), and responses should include IDs for chaining, but this is not verified in the source.
Analyst ratings and price targets for a company.
Company profile: sector, industry, website, description, CEO, employees, headquarters.
Corporate events calendar (earnings, dividends, splits, etc.).
Earnings calendar with upcoming earnings dates and times.
Historical earnings dates and results for a company.
Economic calendar events and releases.
Output schemas completely undocumented. No tool declares what fields it returns (e.g., quote returns price/change/bid/ask/volume/market_cap/beta/EPS, but this is not in a schema). LLMs cannot plan downstream calls or extract data reliably.
| Scored | Grade | Overall | Spec posture | Rubric |
|---|---|---|---|---|
| 2026-09-23 | D | 50 | 2026-07-28+ | v2 |
Full text of an SEC filing for reading.
Financial statements (income statement, balance sheet, cash flow) from SEC filings.
Find a ticker from a company's NAME, or check one you were given: {query, results: [{symbol, name, exchange, asset_class}]}, best match first. CALL THIS BEFORE GUESSING. Every other tool here takes a ticker, and a guess that lands on the wrong listing answers confidently about the wrong company — "Robinhood" is HOOD, but an ETF named after a company often carries a symbol closer to its name than the company's own. The ranking puts an exact ticker first, then a ticker prefix, then a name prefix, then all your words present; an exchange listing outranks an over-the-counter one and a derivative is demoted. Keyless: this is the SEC's own ticker list plus the coin pairs that can be charted, so it answers whatever the reader has connected.
Recent insider trading activity (Form 4 filings) for a company.
Key statistics for a company: shares outstanding, float, insider ownership, institutional ownership, short interest, dividend yield, earnings date.
SEC filings (10-K, 10-Q, 8-K, etc.) for a company.
Index, rate, commodity and crypto levels: the top-of-terminal strip. Returns [{symbol, label, price, change_pct}].
What is happening today, what is trending, what to look at.
Most active, gainers and losers in the market.
Most active, gainers and losers. Filtered for tradeability: warrants, rights and units are excluded, and rows must clear a price and dollar-volume floor. Note that gainers/losers skew small-cap — a percentage screen over the whole market always does. Large names appear on most_active, which ranks by volume.
The reader's own names ("my stocks", "my watchlist").
Full text of a news article where `full_text` is true (the reader's feed already carries it; publishers often refuse automated fetches of their pages).
Insider and institutional ownership information for a company.
Intraday bars for one symbol with RSI-9 and MACD(12,26,9), thinned to at most `points` (default 120, max 400) evenly spaced samples — the last bar is always one of them, and `thinned_every` says how many bars each sample stands for. `bar_count` is the real number behind the samples. Each sample is {t, o, h, l, c, v, rsi_9, macd, macd_signal}, oldest first; the indicator fields are computed on the FULL series, not on the thinned one, so they mean what they would on the chart. For daily history over months or years use `price_history`. IMPORTANT: check `indicators_reliable` before using the indicator values. On a sparse feed an illiquid name's "1-minute" bars can be a handful of prints stretched across days, which computes indicators that look normal and mean nothing. `coverage` and `median_gap_min` say how real the series is. When it is false, describe price only.
Daily price history and returns for a symbol over months or years.
Full quote for one US-listed symbol: price and change, bid/ask, day and 52-week ranges, volume, market cap, valuation multiples, beta, EPS and analyst targets.
Quotes for a basket of symbols.
Breadth statistics for market sectors (advancing/declining issues).
Performance of market sectors.
News articles for a symbol from the reader's feed.
14 of 26 tools have empty input schemas ({}). No parameter type definitions visible in source.
No error handling or recovery guidance. Tools do not document what happens on invalid input (e.g., invalid ticker, missing data, rate limits). LLMs receive no actionable error messages to self-correct.
Vague descriptions on 8 tools (my_board, sector_performance, sector_breadth, market_today, economic_calendar, corporate_calendar). Descriptions under 50 chars or generic ('What is happening today'). LLMs cannot determine when to call these vs. similar tools.
Duplicate tool: 'movers' appears twice (tools #3 and #20) with identical descriptions but different input schemas. LLMs will conflate them, wasting reasoning cycles. Consolidate into one canonical tool.