Fahali

18 detection engines across crypto and US equities/ETFs return read-only market-risk verdicts — every one judged against realized outcomes in a public ledger, misses included.

GitHub

Documentation

Fahali MCP Server

Market-risk intelligence your AI agent can call. Fahali gives agents structured, read-only judgment about markets — direction, confidence, reasoning, expected move, contagion, crash precursors — instead of raw prices they have to interpret blindly.

Informational only — not financial advice. Fahali observes and explains risk. It never tells you or your agent what to buy or sell. No order routing, no path to capital.

What makes it different

Most market APIs hand an LLM data exhaust and hope it reasons well. Fahali hands it a verdict with a receipt:

18 detection engines scan ~840 instruments per cycle across a 2,000+ instrument universe — crypto 24/7 and US equities/ETFs during market hours. Anomaly, capital flow, market regime, dark-pool proxy, whale activity, funding stress, liquidation cascade, tail-dependence/contagion, and 72-hour crash-precursor forecasting.
An adaptive ensemble re-weights those 18 engines every 15 minutes based on which ones have been right lately, per market regime.
Every served claim is judged against realized outcomes and gets a permanent, public replay URL — hits and misses. The full grading methodology is published at fahaliai.com/methodology. Engines without enough resolved outcomes report calibrating — never an invented number.

That last point is the whole thesis: an agent (or the human behind it) can audit any Fahali claim against what the market actually did. Example receipts, misses included, live at app.fahaliai.com.

Connect

Remote MCP (recommended — nothing to install):

Streamable HTTP:  https://mcp.fahaliai.com/mcp
SSE (legacy):     https://mcp.fahaliai.com/sse

Auth: OAuth 2.1 ("Connect with Fahali") or a static key — Authorization: Bearer sk_live_.... Grab a free developer key (50 calls/day) at app.fahaliai.com/developer.

Claude / ChatGPT / Cursor: add the remote MCP URL above as a connector. Listed in the official MCP registry as com.fahaliai/fahali.

SDKs for building agents directly:

npm install fahali      # TypeScript — OpenAI tools format + Vercel AI SDK / LangChain adapters
pip install fahali      # Python — LangChain / CrewAI / LlamaIndex adapters

Tools (25)

Market-intelligence tools are strictly read-only. A few optional agent-memory and custom-alert tools write only to your own agent workspace, never to markets.

Tool	Returns
`fahali_get_market_verdict`	Per-symbol verdict: direction, confidence, reasoning chain, expected move, horizon
`fahali_get_72h_forecast`	Probabilistic 72h forecast: crash/neutral/pump probabilities, expected return, uncertainty cone
`fahali_get_portfolio_risk`	Portfolio risk: score, VaR, drawdown, per-position breakdown
`fahali_get_flash_crash_risk`	Flash-crash precursor signals
`fahali_get_whale_activity`	Large-order / whale flow
`fahali_get_dark_pool_activity`	Off-tape absorption (proxy)
`fahali_get_contagion_map`	Cross-asset tail-dependence / correlated clusters
`fahali_get_capital_flow`	Net flow, institutional vs retail, per symbol
`fahali_get_market_regime`	Current regime read (HMM-based)
`fahali_get_track_record_scorecard`	The judged record with per-horizon base rates
`fahali_run_shock_test`	Stress-test a portfolio against a scenario
…	14 more — market snapshot, sentiment, correlation matrix, case studies, engine status, briefing, and agent-workspace tools

Full tool schemas: mcp.fahaliai.com.

Pricing

Free developer key (50 calls/day) for exploration. Agent lanes from $49/mo (10k calls) → $199 (100k) → $999 (1M). Human app tiers from $19/mo at fahaliai.com.