canli-mcp

canli-mcp

Every Canli Capital MCP server in one: 315 finance tools for Claude, Cursor or any MCP client.

• Quant finance: 235 tools covering performance and risk, options and exotics, fixed income, portfolios, econometrics, indicators, and a library of 399 strategy sleeves with multiple-testing corrections.
• Backtest validation: deflated Sharpe, overfitting probability, data-snooping tests, leakage checks and placebo tests, run on your machine.
• Point-in-time SEC fundamentals.
• An analyst's filings, and a check on the answer: financial statements exactly as filed (several years joined, restatements and splits shown beside the first-reported figures), the table in any filing that holds a figure (followed through earlier filings), earnings releases with their guidance as ranges and last quarter's guidance set against the results, and verify_claims, which checks a draft's figures against the filings: confirmed, contradicted (with the filed figure and what the wrong one matches) or not reported.
• Markets from the sources, no key: screens of every US-listed company, company reports, event studies, SEC filings and their sections, insider trades, 13F and 5% holders, fund and ETF portfolios, index members, short interest, off-exchange volume, fails to deliver, futures positioning, executive pay, revenue by segment, options chains and volatility surfaces, the VIX futures curve, Fed decisions and projections, the economic calendar, Treasury yields and auctions, the federal debt, FRED, World Bank and OECD data, and prices for stocks, currencies, futures, indices and crypto.
• Beyond the US, from the publishers: European companies' annual reports as they filed them (ESEF), central-bank rates (SOFR, the euro short-term rate, SONIA, the ECB's and the Bank of England's rates, and about forty policy rates the BIS collects), what Polymarket and Kalshi traders pay for an outcome, and DeFi value locked, fees and stablecoin supply from DefiLlama.
• Canli Capital's open research record.
• Factor backtests without lookahead.
• Alpaca paper trading behind pre-trade checks.

Each pack is a published, separately tested server. This package loads them into one process and puts three tools in front of them.

npx -y canli-mcp

Claude Code:

claude mcp add canli -- npx -y canli-mcp

Claude Desktop, Cursor or any MCP client:

{ "mcpServers": { "canli": { "command": "npx", "args": ["-y", "canli-mcp"] } } }

Hosted, no install: add https://canlicapital.com/mcp/canli as a connector in claude.ai, ChatGPT or any client that takes a URL (claude mcp add --transport http canli https://canlicapital.com/mcp/canli). It runs this release with the quant, validation, fundamentals, research, backtest and markets packs. File arguments ($file), prices and paper trading need the local server: files there would be the server's disk, the keyless price source is for personal use, and paper trading needs your own broker keys. The hosted server keeps nothing between requests, so chain calls inside one run_tool batch, and each caller gets 60 requests a minute.

Context and startup

Measured on Node 24 on 2026-10-15 with bench/context.py (o200k tokens of the tool objects a client receives, plus the server instructions; all seven packs enabled with CANLI_PACKS=all; start-up is the median time from launch to the tool list, on a machine running other work; bench/context-2026-10-15-0.6.0.json):

	seven servers separately	canli-mcp
initial tool list and instructions	39,962 tokens	1,160 tokens
processes	7	1
start-up	71-100 ms each	86 ms

The four tools this release adds cost the combined server nothing up front; run separately, the markets server's list grew by 1,915 tokens.

• Three tools in context. The model searches with find_tool, reads one schema with describe_tool, and calls anything with run_tool. On 48 everyday requests written after the last change to the search and never tuned on, the right tool comes first 46 times and is in the top three 47 times (0.2.1: 25 and 29). For the eight tools added in 0.5.0 and 0.6.0, 41 requests written before measuring found the right tool first 36 times and in the top three 38 times; two words they used were then added to the index (38 and 40). test/search.test.mjs holds these and 176 more.
• Nothing loads until it is used. Start-up reads a prebuilt index of all 315 tools; a pack's code loads the first time one of its tools runs.
• Data by file, not by pasting. Any argument can be {"$file": "prices.csv", "column": "close"} (or "columns": ["SPY", "TLT"] or "all"). For 1,000 prices that is 23 tokens instead of 4,893, and the model cannot drop a number while copying.
• Several calls in one round trip. run_tool takes calls (up to 25). A call can use an earlier call's output with {"$result": 0, "path": "rows", "pick": [0, 1]}, so intermediate data never passes through the model, and return: "last" sends back only the final answer. Labeling 1,000 prices and then weighting the labels took 72 tokens of tool input this way, against 31,917 to read the labels and type them into the next call.
• Shorter numbers when they are enough. digits: 5 rounds every result.

Workflows as prompts

Clients that show MCP prompts (Claude Desktop and Claude Code list them as slash commands) offer ten workflows. Each one runs a single batched run_tool call and says what to write from the results:

prompt	asks for	runs
company_brief	a ticker	company report, 8-K news with press-release headlines, the latest results against their guidance, next earnings date, insider trades, short interest, options-implied move
earnings_preview	a ticker	earnings date and timing, what the company guided for the quarter and whether it met its last guidance, the move options price in, past reactions to results (event study), recent results headlines, revenue by segment
financial_history	a ticker	four years of statements as filed, with restatements and splits shown
fact_check	a ticker and a draft	every reported figure in the draft checked against the filings, the draft returned corrected
macro_snapshot	nothing	the Fed's latest decision and projections, what prediction markets price for its next decisions, SOFR against the Fed's range, the next two weeks' key releases, CPI, core PCE, unemployment, the Treasury curve, the VIX futures curve
global_rates	nothing	the Fed's, ECB's, Bank of England's and Bank of Japan's policy rates, SOFR, the euro short-term rate and SONIA beside them, the Treasury curve
european_company	a company's name or LEI	its latest ESEF annual report: revenue, profit, EPS, assets, equity, cash flows and dividends beside the prior years, linked to the report
defi_snapshot	a protocol (optional)	stablecoin supply and its change, value locked on Ethereum and Solana, the protocol's value locked, fees and revenue
ownership	a ticker	5% holders and activists, insider buying and selling, short interest
backtest_check	a returns file	deflated Sharpe for the variants tried, the track record needed, then leakage and placebo plans

A prompt is offered only when every tool it names is in an enabled pack (CANLI_PACKS, CANLI_OFFLINE), and test/prompts.test.mjs checks each call against its tool's own schema. Prompts are requested separately from tools/list. Their context cost depends on how the client includes prompt and resource metadata.

The packs also bring their own 25 prompts, so the full server offers 35. Their names identify the pack, such as fundamentals.known_on_date, quant.audit_backtest, validation.stress_my_strategy and research.literature_review. They route tool calls through run_tool, preserve the source pack's prompt argument checks, and forward its completion helpers.

The 15 static resources and nine resource templates include the 399 sleeve specifications, reference methods, exact input/output schemas, concept mappings, limits and client examples. Pack resource schemes keep their names separate: canli-quant://, canli-validation://, canli-fundamentals:// and canli-research://. For example, read canli-fundamentals://schemas/cross_section. Original client examples retain each pack's local or hosted connection. Disabled packs have no guides; listing guides reads the prebuilt metadata and loads no pack code. The guide handlers load only when a prompt, resource or completion is used.

Results drawn in the chat

Clients that draw MCP Apps (the extension of 2026-01-26) show each run_tool result as a page beside the answer: tables with the figures aligned, statements as filed with their scale, checked claims under their sentence with a verdict for each figure, last quarter's guidance as a range with the actual result marked, and a line chart for prices or any series of dates. A result's grouped figures (a stock's market data, the next earnings date, a protocol's fees) are drawn as labelled figures, the Treasury curve over its maturities, a futures curve by expiration, a policy rate as steps between its decisions, and a market's probability as a percentage. The page is the resource ui://canli/view, one self-contained file with no network access and no external scripts, drawn in the client's sandbox; every string from a result is escaped. What the model reads does not change, and clients without MCP Apps ignore the page.

FEATURES.md maps the supported finance workflows, evidence checks and remaining gaps. For Claude Desktop, the GitHub release also carries a .mcpb extension with its dependencies and GitHub build attestation. Its manifest lets you choose enabled packs and offline mode.

An agent, end to end

The same model (gpt-5.4-mini) answered seven tasks twice each:

• the Sharpe ratio, drawdown and value at risk of a pasted price file;
• a deflated Sharpe from summary numbers;
• a Black-Scholes price;
• a 399-sleeve tournament on three assets;
• Apple's first-reported 2019 EPS.

Answers were graded against the tools' own values (bench/agent-eval.mjs; results in bench/agent-eval-2026-10-08.json):

	correct	tokens per task	turns	seconds
the six servers, installed separately	11 of 14	63,984	4.1	22.6
canli-mcp, first version	10 of 14	43,043	5.8	9.5
canli-mcp, after fixing what the runs showed	14 of 14	27,267	4.2	7.2

The fixes came from reading the runs:

• The model narrowed find_tool to one pack and missed the tool that took its inputs, so the pack filter is gone.
• It passed made-up two-point series to a tool that wanted returns, so the quant deflated and probabilistic Sharpe tools now take summary numbers and refuse too little data.
• It put select and return inside a tool's arguments, so run_tool moves them out.
• It asked select for fields that do not exist, so the reply now lists the fields that do.

This is one small model with two runs per task, so read the differences as large or small, not as exact rates.

Privacy

Everything runs on your machine. canli://privacy states, per pack, what leaves it:

pack	network	disk
quant	none	none
validation	none (private local mode)	reads a returns file only when you pass its path
fundamentals	public SEC data from canlicapital.com (the company and measure you ask for)	caches that public data in ~/.cache/canli-fundamentals (CANLI_CACHE_DIR="" keeps nothing)
research	public research pages from canlicapital.com (your search words)	none
backtest	public SEC data, as fundamentals; your prices and signals stay local	your trial ledgers in ~/.canli/ledgers
markets	SEC EDGAR and the SEC's adviser database, the US Treasury, FRED, FINRA, the CFTC, the Federal Reserve and the New York Fed, BLS, BEA, the Census Bureau, the OECD, the World Bank, Cboe, filings.xbrl.org (European filings), the ECB, the Bank of England, the BIS, Polymarket, Kalshi and DefiLlama directly (the company, form, dates, search words, rate or series you ask for); Yahoo Finance's public chart data for prices, or for US stocks Alpaca or Tiingo with your key	none
paper	Alpaca's paper API only, with your paper keys	a hash-chained order log in ~/.canli

CANLI_OFFLINE=1 enables only the packs with no network, and no other pack's code is even loaded. test/privacy.test.mjs runs the server that way with fetch and every network module made to throw, and the quant and validation tools still answer. The server itself has no network, write or logging code (also tested). Opt-in receipts (receipt: true) carry hashes, never data.

Configuration

variable	effect
CANLI_PACKS	Packs to enable: quant,validation,fundamentals,research,backtest,markets,paper or all. Default: all but paper.
CANLI_OFFLINE	1 keeps only quant and validation.
ALPACA_PAPER_KEY_ID, ALPACA_PAPER_SECRET_KEY	Paper keys (starting PK); setting them enables the paper pack. Live keys are refused.
CANLI_HOME	Where the paper pack keeps its limits file, kill switch and order log; default ~/.canli.

Order-sending tools never run inside a batch: preview first, then send on its own.

Packs and what each is checked against

pack	package	tools	tested against
quant	canli-quant-mcp	235	QuantLib, statsmodels, arch, TA-Lib, scipy, pandas: 465 reference cases
validation	canli-validation-mcp	17	the Null Zoo benchmark of 160,000 simulated searches
fundamentals	canli-fundamentals-mcp	7	SEC XBRL filings
research	canli-research-mcp	6	the published research record and its hash chain
backtest	canli-backtest-mcp	3	point-in-time SEC data
markets	canli-markets-mcp	43	replayed responses from the SEC, Treasury, FRED, FINRA, CFTC, the Fed, OECD, World Bank, Cboe, ECB, Bank of England, BIS, filings.xbrl.org, Polymarket, Kalshi and DefiLlama: a 13F's total equals its cover page; ESEF figures equal the same companies' Form 20-F at the SEC; central-bank rates agree with the BIS
paper	canli-paper-trading-mcp	4	a simulated Alpaca paper API

Two tools differ from their standalone servers. Validation's stress_test is strategy_stress_test here, because the quant pack has a portfolio stress_test. Validation's get_key and service_status are left out, because validation runs locally here. canli://packs lists both.

Development

npm ci && npm test
npm run build-index        # after updating a pack; a test fails if the index and the packs differ
node bench/robustness.mjs  # every tool, offline, with hostile arguments: crashes and leaks reported

MIT licence.