# VARRD — The Governed Live Edge Layer

> Statistically validated trading edges, monitored in real time. Full transparency on every edge — the formula, the methodology, the performance, and the audit trail.

## What VARRD Is

VARRD was built by a team from one of the most successful derivatives trading firms in Chicago history, alongside Princeton graduates and AI engineers. The firm's process for taking a raw trading idea and turning it into a validated, deployable system — the same process that produced decades of real returns — is what VARRD is built on. We took that methodology and made it accessible to anyone, with AI handling the heavy lifting and a purpose-built engine enforcing every rule.

VARRD maintains a growing library of statistically validated trading edges running 24/7 against live market data across futures, equities, and crypto. When an edge fires, you get market, direction, entry, stop, target, hold period, and the complete methodology behind it. Every edge in the library survived rigorous statistical testing with proper guardrails — Bonferroni correction, K-tracking, lookahead verification, and out-of-sample validation.

You can also test your own ideas. Describe any trading hypothesis in plain language and VARRD loads real market data, charts the pattern, runs statistical tests, and tells you if there's an edge. The AI generates hypotheses. A purpose-built engine does all the math. The AI never calculates statistics or fabricates results — every number comes from deterministic computation on real data.

## Questions VARRD Can Answer

These are real things you can ask. VARRD loads the data, builds the formula, tests it, and gives you a clear verdict.

- "Is there an edge to going long Starbucks on their Red Cup Day?"
- "What happens to gold futures when yen and 10-year bonds make 20-day new highs on the same day?"
- "Take a look at crude oil on a 240-minute timeframe right now — diagnose how it's acting and let's find every time it's happened in the past."
- "Does it test out to short fast food companies during Lent?"
- "When corn and soybeans diverge by more than 2 standard deviations, is there a mean reversion trade?"
- "What happens to Bitcoin after a halving event?"
- "Use the ELROND expert council on natural gas" (AI specialists each return calibrated formulas)

## The Edge Library

The headline product. A live feed of validated edges across futures, equities, and crypto.

**What's in it:** Edges discovered through deep statistical testing — event studies, backtests with stops, multi-market validation. Each edge has a formula written in a domain-specific language built to quantify market behaviors, the full discovery story, per-horizon results, and ongoing post-discovery performance tracking.

**How it works:**
- Edges are scanned against live data every 15 minutes
- FIRING = signal confirmed, enter at next bar's open
- PENDING = bar hasn't closed yet, may fire when it does
- ACTIVE = already in a trade from a previous signal

**Pricing (MCP / CLI):**
- FREE: See which markets have edges firing, pending, or in active trades
- $0.50: Unlock direction, win rate, expected value, stop/target, entry date for ALL active edges
- $1 per edge: Full methodology — formula, discovery story, performance analytics, per-horizon p-values, interactive chart
- $5 for all edges: Everything on every edge at full depth
- Section drill-downs (free after purchase): setup code, horizon table, full analytics (SQN, Kelly, Monte Carlo, drawdown, regime analysis, edge decay), every historical signal, interactive chart link

**Pricing (Web App):**
- $2 for the snapshot view (direction, entries, exits, live performance)
- $5 for full depth (formula, setup code, testing history, load into your own session)

## How to Connect

**MCP (Claude Desktop, Cursor, any MCP client):**
```json
{ "mcpServers": { "varrd": { "url": "https://app.varrd.com/mcp" } } }
```
Works immediately with anonymous access. To get $2 in free credits:
1. Sign up at https://app.varrd.com (verify email)
2. Click avatar (top right) → Manage Agents
3. Copy the MCP config JSON shown there (includes your auth token)
4. Paste into your MCP client config and restart

**CLI:**
```
pip install varrd
varrd edges # see what's firing (free)
varrd edges --depth 1 # stats + trade levels ($0.50)
varrd research "your idea here"
```

**Web App:** https://app.varrd.com — sign up, get $2 free, same capabilities with a visual interface.

## MCP Tools

| Tool | What it does | Cost |
|------|-------------|------|
| `varrd_edges` | Browse the validated edge library — filter by market, direction, timeframe, asset class | Free at depth 0, $0.50/$1/$5 at deeper tiers |
| `varrd_ai` | Multi-turn research — test any trading idea with real data | ~$0.25/session |
| `autonomous_varrd_ai` | Fully autonomous edge discovery — give it a topic, it finds edges | ~$1/idea |
| `search` | Search your saved strategies by keyword or concept | Free |
| `get_hypothesis` | Full detail on a specific strategy | Free |
| `check_balance` | Credit balance + auto-detects completed payments | Free |
| `buy_credits` | Add credits via Stripe Checkout ($5 min) or USDC on Base | Free to call |
| `get_briefed` | Personalized market news tied to your edge library | Credits |
| `reset_session` | Kill a stuck research session | Free |

## What's Impossible on VARRD

These aren't guidelines. They're hard stops enforced at the infrastructure level — the same whether you're a human clicking buttons or an AI agent running at 3am.

- **You cannot cherry-pick results.** Every test you run is counted and raises the significance bar for the next one. Test 50 variations of an idea? The system knows, and adjusts accordingly.
- **You cannot peek at out-of-sample data.** Once it's used, it's locked permanently. No re-running. No optimizing after the fact. One shot.
- **You cannot use future data in a formula.** The system detects lookahead bias automatically and verifies signals reproduce on truncated historical data.
- **You cannot fabricate statistics.** The AI interprets results but never computes them. Every number comes from a deterministic engine running on real market data. The AI and the math are architecturally separated.
- **You cannot test the same thing twice to fish for a better number.** Every test is fingerprinted. Duplicates are caught and blocked.
- **You cannot disguise an old idea as a new one.** Vectorized embeddings detect when a hypothesis overlaps with something already tested — even if you reword it completely.

## Full Transparency

Every edge shows you everything:
- **How it was found** — the discovery story, the hypothesis, the market structure theory behind it
- **The formula** — written in a domain-specific language, fully auditable
- **How it was tested** — per-horizon results, K value, p-values, lookahead verification
- **How it's performing now** — post-discovery performance tracked separately from in-sample, edge decay over time, regime analysis, Monte Carlo simulations
- **Interactive view** — browser link showing the chart, every signal, equity curve, and full stats (shareable, no account needed)

## Data Coverage

- **Futures (CME):** Energies, metals, grains, currencies, indices, bonds — hourly through weekly, back to 1985
- **Equities:** Any US stock or ETF — hourly through weekly
- **Crypto (Binance):** BTC, ETH, SOL and more — hourly through weekly

## Credits and Payment

- Sign up at app.varrd.com → $2 free credits (requires email verification)
- MCP/CLI agents start at $0 — add credits via `buy_credits` tool (returns a Stripe Checkout link, $5 minimum)
- Crypto: `buy_credits(payment_method='crypto')` for USDC on Base — fully autonomous, no human needed
- After payment, call `check_balance` to confirm credits were added
- Research costs ~$0.25 per session. Autonomous discovery ~$1 per idea.

## The Bottom Line

VARRD is where trading ideas go to be tested honestly. The math is real, the guardrails are enforced whether you want them to be or not, and every edge in the library earned its place through genuine statistical validation. When an idea doesn't have an edge, that's a valuable result too — knowing what doesn't work is as important as knowing what does.

**MCP:** https://app.varrd.com/mcp
**Web:** https://app.varrd.com
**CLI:** `pip install varrd`
**GitHub:** https://github.com/augiemazza/varrd
**Contact:** management@varrd.com