Every market AI claims to be right. Fahali proves when it's wrong: market-risk calls with explicit probabilities and deadlines, publicly graded against what the market actually did — permanent replay links, misses kept forever. 18 detection agents across crypto + US equities/ETFs watch what you hold and explain risk in plain language. Humans use the app; AI agents call it over MCP (Claude, ChatGPT, Cursor) or the fahali SDKs. Read-only — observation, not advice.

Hey Product Hunt 👋 — solo founder here. I built Fahali out of a simple frustration: every trading tool, signal group, and "AI analyst" on the internet claims to be right, and none of them will show you their misses. The industry runs on survivorship bias and deleted tweets. So I built the opposite. Fahali makes market-risk calls with explicit probabilities and deadlines — and then grades itself in public. Every call gets a permanent replay page showing what we said, when we said it, and what the market actually did. We can't edit them. We can't delete them. Two receipts from the last 24 hours, so you don't have to take my word: ✅ A hit — flagged MMT as institutional distribution; it dropped ~12% by judgment: app.fahaliai.com/replay/committee_verdict-MMTUSDT-1783834167707 ❌ A miss — read SKL as accumulation; the market went ~16% the other way. It stays on the record forever: app.fahaliai.com/replay/committee_verdict-SKLUSDC-1783747660088 That second link is the product. Anyone can be right sometimes; the question is whether a tool is honest enough to be audited. Under the hood: 18 detection engines (order flow, dark-pool proxies, whale activity, funding stress, liquidation cascades, regime shifts, contagion, 72h crash precursors) across crypto + US equities/ETFs, fused by an adaptive ensemble that re-weights engines every 15 minutes based on which ones have actually been right lately. The judging methodology is fully public: fahaliai.com/methodology — including per-horizon base rates, so nobody (including us) can quote a misleading number. Two ways to use it: 🧍 Humans: connect your exchange read-only and it watches what you actually hold, flagging risk building underneath in plain English. 🤖 AI agents: it's a remote MCP server (works in Claude, ChatGPT, Cursor — official registry: com.fahaliai/fahali) plus npm i fahali / pip install fahali. Your agent asks "what's the risk on X?" and gets a structured, graded answer instead of raw candles. What Fahali is NOT: financial advice, a signal group, or an execution bot. It's read-only, it never says buy/sell, and we refuse to publish a single "accuracy %" — the ledger is the answer, and some engines are honestly still calibrating. For PH: the free developer key (50 calls/day, no card) is live today, and I'm personally onboarding the first 20 founding members at $19/mo — 15-min call with me, price locked forever. Roast it. Genuinely. I publish my misses — I can take yours. I'll be here all day answering everything. 🦁

The fact that you keep misses forever is honestly really cool, most platforms just bury their bad calls. One thing i'd love to see is some kind of portfolio-level risk score that updates as the 18 agents flag stuff, kind of a single number i can glance at instead of reading every individual alert. would make it way easier to decide when something actually needs my attention.

Love that missed calls stay on the permanent record, that's a rare level of accountability for an AI tool. The plain-language risk explanations actually read like something a thoughtful friend would say, not jargon soup.

Fahali

The market AI that grades its own calls — misses included

The market AI that grades its own calls — misses included