MCP SERVER · OFFICIAL PROJECTPull FinBrain's price forecasts and alternative datasets for a ticker — congress trades, insider filings, lobbying, patents, contract awards, Reddit mentions — or screen the same signal across a whole market.
LABSCO SUMMARY
The predictions are the headline; the alternative data is what is genuinely hard to find in one place — congressional trades with the disclosure lag preserved, lobbying filings, patents by assignee, contract awards, Reddit mentions and app-store ratings. The paired design pays off in use: every per-ticker dataset has a screener twin, so moving between "what about this company" and "who else looks like this" does not mean changing tools. The CIK carried on the regulatory datasets is the small detail that makes any of it joinable to EDGAR.
A local server over the FinBrain v2 API that exposes its datasets in pairs: one tool per ticker and a screener twin covering the same signal across a market or region. Coverage runs from machine-learning price predictions to alternative data that rarely reaches an assistant — congressional trading, corporate lobbying, USPTO patents and US government contract awards.
Python 3.10 or newer, published on PyPI as finbrain-mcp and launched over stdio. It runs locally against your own FinBrain API key in FINBRAIN_API_KEY, with nothing proxied through a third party, and reaches FinBrain through the official finbrain-python SDK on the v2 API.
One command plus a key — pip install --upgrade finbrain-mcp, then supply credentials
USE CASES
The forecast, the sentiment series and the insider filings for a single ticker, each returned in the same normalised shape.
Screen sentiment, analyst ratings and insider trading across a market or region instead of checking tickers one at a time.
House and Senate trades with both the trade date and the disclosure date, plus the lobbying filings behind them.
Headcount and follower trend, granted patents by assignee, and the government contracts it has been awarded.
The put/call ratio with its underlying counts, and Reddit mentions broken down by subreddit over time.
The CIK on each regulatory row is the join key to EDGAR filings or a security master — keep it as text, the leading zeros matter.
Counted once with only startup setup. Other versions, modes or optional credentials may expose a different list.
analyst_ratings_by_tickerNormalized analyst ratings:
{
format: "json",
ticker, name,
series: [{date, rating_type, institution, signal,
target_price_from, target_price_to, target_price_raw}, ...],
series_count, series_total
}
CSV returns the sliced series.
app_ratings_by_tickerNormalized app ratings with paging over series.
JSON:
{
format: "json",
ticker, name,
series: [{date, play_store_score, play_store_ratings_count, app_store_score,
app_store_ratings_count, play_store_install_count}, ...],
series_count, series_total
}
CSV: CSV text of the sliced series.
available_marketsList available markets (e.g., 'S&P 500').
available_regionsList markets grouped by region. Use region codes for filtering in other endpoints.
available_tickersList available tickers for a dataset.
corporate_lobbying_by_tickerNormalized corporate lobbying filings:
{
format: "json",
ticker, name,
series: [{date, filing_uuid, filing_year, quarter,
client_name, registrant_name,
income, expenses, issue_codes, government_entities}, ...],
series_count, series_total
}
CSV returns the sliced series.
government_contracts_by_tickerU.S. government contract awards for a single ticker:
{
format: "json",
ticker, name,
series: [{award_id, award_amount, award_type, awarding_agency,
awarding_sub_agency, recipient_name, start_date, end_date,
description, naics_code, naics_description,
contract_award_type}, ...],
series_count, series_total
}
CSV returns the sliced series.
healthBasic server health & version info. Tries to resolve the API key and construct the SDK client (no network call), then returns versions.
house_trades_by_tickerNormalized US House trades:
{
format: "json",
ticker, name,
series: [{date, representative, trade_type,
amount_min, amount_max, amount_exact, amount_raw,
disclosure_date, owner, amount_as_filed, amount_flag}, ...],
series_count, series_total
}
date is the transaction date; disclosure_date is when the trade was
publicly disclosed in the member's periodic transaction report. The gap
between them is the reporting lag.
owner is the beneficial owner of the traded account: "SELF", "SP"
(spouse), "DC" (dependent child), "JT" (joint), or an account code;
blank House filings report as "SELF". owner and disclosure_date are
nullable, though rare — historical rows were backfilled upstream.
amount_raw is the amount string as served by the API, normalized to a
statutory STOCK Act bracket when the filed string is an unambiguous
variant of one; amount_as_filed then preserves the original filing.
amount_flag is "review" or "ambiguous" when the filed amount could not
be safely normalized (amount_raw keeps the filed string, or "Unknown");
both are null on all other rows.
CSV returns the sliced series.
insider_transactions_by_tickerNormalized insider transactions with paging over series.
JSON:
{
format: "json",
ticker, name,
series: [{date, insider_name, relationship, transaction_type,
price, shares, usd_value, total_shares,
sec_form4_date, sec_form4_link}, ...],
series_count, series_total
}
CSV: sliced series as CSV.
linkedin_metrics_by_tickerNormalized LinkedIn metrics:
JSON:
{
format: "json",
ticker, name,
series: [{date, employee_count, followers_count}, ...], # paged
series_count, series_total
}
CSV:
CSV text of the sliced series.
news_by_tickerReturns recent news articles for a ticker: { format: "json", ticker, name, series: [{date, headline, source, url}, ...] (paged), series_count, series_total } If format="csv", returns CSV of the sliced series.
news_sentiment_by_tickerReturns normalized sentiment: { format: "json", ticker, name, series: [{date, score}, ...] (paged), series_count, series_total } If format="csv", returns CSV of the sliced series.
options_put_callNormalized options put/call time series. JSON: { format: "json", ticker, name, series: [{date, put_call_ratio, call_count, put_count}, ...], # paged series_count, series_total } CSV: CSV text of the sliced series.
patent_filings_by_tickerUSPTO granted patents mapped to a single ticker by corporate assignee:
{
format: "json",
ticker, name,
series: [{patent_id, patent_date, title, type, kind, num_claims,
num_cited_by, assignee_organization, assignee_type,
application_filing_date, filing_to_grant_days, inventors,
num_inventors, cpc_sections, cpc_subsections,
primary_cpc_section}, ...],
series_count, series_total
}
CSV returns the sliced series.
predictions_by_marketScreener-based market predictions (flat rows with expected_* percentages). Filters by market name or region.
predictions_by_tickerNormalized ticker prediction with a time series: { ticker, name, type, last_update, expected_short/mid/long, series: [{date, mid, low, high}, ...], }
recent_analyst_ratingsGet the most recent analyst ratings across all tracked stocks. Returns rows with ticker, name, date, institution, rating_type, signal, target_price.
recent_newsGet the most recent news articles across all tracked stocks. Returns rows with ticker, name, date, headline, source, url.
reddit_mentions_by_tickerReddit mention counts across subreddits for a single ticker:
{
format: "json",
ticker, name,
series: [{date, subreddit, mentions}, ...],
series_count, series_total
}
CSV returns the sliced series.
screener_analyst_ratingsScreen analyst ratings across tickers. Returns rows with ticker, name, date, institution, rating_type, signal, target_price.
screener_app_ratingsScreen app ratings across tickers by market or region. Returns rows with ticker, name, date, app_store_score, play_store_score. Either market or region is required.
screener_government_contractsScreen U.S. government contract awards across all tickers. Returns rows with ticker, name, award_id, award_amount, recipient_name, start_date, awarding_agency, naics_description, plus a summary with aggregate stats (total_contracts, total_tickers, total_value).
screener_house_tradesScreen House of Representatives trades across all tickers. Returns rows with ticker, name, date, politician, trade_type, amount, disclosure_date (public disclosure date) and owner (beneficial owner of the account: "SELF", "SP", "DC", "JT", or an account code). owner and disclosure_date are nullable, though rare — historical rows were backfilled upstream.
screener_insider_tradingScreen insider trades across all tickers. Returns rows with ticker, name, date, insider_name, relationship, transaction_type, shares, total_value.
screener_linkedinScreen LinkedIn data across tickers by market or region. Returns rows with ticker, name, date, employee_count, followers_count, job_count. Either market or region is required.
screener_newsScreen news across tickers. Returns rows with ticker, name, date, headline, source, url.
screener_patent_filingsScreen USPTO granted patents across all tickers. Returns rows with ticker, name, patent_id, patent_date, title, type, assignee_organization, primary_cpc_section, num_claims, plus a summary with aggregate stats (total_patents, total_tickers, top_cpc_sections).
screener_put_call_ratioScreen put/call ratio across tickers. Returns rows with ticker, name, date, put_call_ratio, call_count, put_count.
screener_reddit_mentionsScreen Reddit mentions across tickers. Returns rows with ticker, name, date, total_mentions, subreddits, plus a summary with aggregate stats (top_mentioned, subreddit_names, etc.).
screener_senate_tradesScreen Senate trades across all tickers. Returns rows with ticker, name, date, politician, trade_type, amount, disclosure_date (public disclosure date) and owner (beneficial owner of the account: "SELF", "SP", "DC", "JT", an account code, or "UNKNOWN" for blank Senate filings). owner and disclosure_date are nullable, though rare — historical rows were backfilled upstream.
screener_sentimentScreen sentiment across tickers by market or region. Returns rows with ticker, name, date, score. Either market or region is required.
senate_trades_by_tickerNormalized US Senate trades:
{
format: "json",
ticker, name,
series: [{date, senator, trade_type,
amount_min, amount_max, amount_exact, amount_raw,
disclosure_date, owner, amount_as_filed, amount_flag}, ...],
series_count, series_total
}
date is the transaction date; disclosure_date is when the trade was
publicly disclosed in the member's periodic transaction report. The gap
between them is the reporting lag.
owner is the beneficial owner of the traded account: "SELF", "SP"
(spouse), "DC" (dependent child), "JT" (joint), an account code, or
"UNKNOWN" for blank Senate filings. owner and disclosure_date are
nullable, though rare — historical rows were backfilled upstream.
amount_raw is the amount string as served by the API, normalized to a
statutory STOCK Act bracket when the filed string is an unambiguous
variant of one; amount_as_filed then preserves the original filing.
amount_flag is "review" or "ambiguous" when the filed amount could not
be safely normalized (amount_raw keeps the filed string, or "Unknown");
both are null on all other rows.
CSV returns the sliced series.
BEFORE YOU INSTALL
33 tools when we connected.
FINBRAIN_API_KEY in your environment
All 3 documented setups run the server on your machine.
CONTEXT COST
Rated by the size of the 33 tool definitions that ride along on every turn, not by how many there are
INSTALLATION
pip install --upgrade finbrain-mcp{
"mcpServers": {
"finbrain": {
"command": "finbrain-mcp",
"env": {
"FINBRAIN_API_KEY": "<YOUR_FINBRAIN_API_KEY>"
}
}
}
}{
"mcpServers": {
"finbrain": {
"command": "docker",
"args": [
"run",
"-i",
"--rm",
"finbrain-mcp:latest"
],
"env": {
"FINBRAIN_API_KEY": "<YOUR_FINBRAIN_API_KEY>"
}
}
}
}{
"servers": {
"finbrain": {
"command": "finbrain-mcp",
"env": {
"FINBRAIN_API_KEY": "<YOUR_FINBRAIN_API_KEY>"
}
}
}
}MCP servers
TradeOS-AI
kevinlin49361128-stack
alraico
Athe-kunal