Inputs
The agent sends the setup it is considering. Leroy does not need a candle dump or a giant prompt.
Command
Leroy CLI
$ leroy evaluate ABNBMarket evidence from the command line.
Install Leroy beside the agent or workflow you already use. See where current research signals agree, then evaluate a ticker by default as a buy case or choose another side explicitly.
No card required. Free accounts include 100 lookups each month.
From request to response.
Scroll through the five passes behind a CLI answer. Every step adds context before Leroy returns the evidence.
CLI
Assess Market
The current market frame adds context around the setup: volume, benchmarks, breadth, and volatility.
Volume
1.38× avg
SPY
+0.42%
VTI
+0.31%
Breadth
61% advancing
Volatility
Normal
Read Strategy Signals
Each supported strategy reports a signal so the agent can see agreement, disagreement, and neutral signals.
Opening Range 30
+0.82
Donchian 20
+0.67
Relative Strength
+0.41
VWAP
0.00
RSI 14
+0.04
Bollinger
−0.08
Find Historical Matches
Leroy searches the verified record for comparable setups and measures what followed them.
AAPL · Buy · 15 minutes
184 comparable historical setups
Average Return
+3.84%
Win Rate
63%
Profit Factor
1.42
Return Evidence
The agent receives the evidence, assumptions, and boundary in one deterministic response.
Possible Return
EvaluatedAAPL · buy
15-minute historical average
+3.84%
Samples
184
Win Rate
63%
Profit Factor
1.42
Agent Wake
Set a rule. Wake your agent.
Tell Leroy what to watch. When a fresh live setup meets the rule, Leroy wakes your trading agent and sends the evidence it needs to decide what happens next.
The delivery is a signed HTTPS webhook with current quote context, signal confluence, match strength, historical outcomes, provenance, and an explicit research-only boundary. Leroy never places an order.
Create a Wake Rule01
Set the rule
Choose a ticker, side, confluence, evidence strength, and outcome threshold.
02
Leroy watches
Only fresh frames during the regular market session can trigger a wake.
03
Your agent decides
Your endpoint receives a signed evidence packet and remains responsible for any action.
The operating loop
A rule is durable. A wake is event-driven. The payload is auditable.
POST https://agent.example.com/leroy/wake
X-Leroy-Event: evidence.setup
X-Leroy-Signature: v1=...
{ "event": "evidence.setup",
"trigger": { "side": "buy", "active_signal_count": 7,
"match_strength": 4, "horizon_minutes": 15 },
"research_boundary": { "live_order": false } }Start with one question.
The default response is compact enough for a person to read and structured enough for an agent to pass along.
leroy evaluate ABNBBUY evidence
Evidence Strength
ABNB@ $191.13
Signal Pattern
50signal-pattern observations
Strategy mix
41 evaluatedBuy 18Neutral 14Sell 9
Observed Outcomes
15 Min+0.84%62% win rate · lift +8.40
30 Min+2.46%64% win rate · lift +12.50
60 Min+3.18%68% win rate · lift +18.10
Commands
Connect once, find current signal leaders, and inspect historical evidence.
leroy connectValidate the connection and save an API key locally with owner-only permissions.
leroy evaluate SYMBOL [--side buy|sell|both]Evaluate the buy case by default, or use --side to evaluate the sell or both-sided case against current market context and Market Memory.
leroy leaders [buy|sell|both] [--limit 1-10]Rank current quote-fresh symbols by observed active signal agreement in one scanner snapshot. This costs one lookup and does not recommend or forecast a trade.
leroy helpPrint the local command summary and a link to this reference.
Flags
--setupInclude the current market setup, including benchmark, volume, session return, VWAP distance, and ATR when available.
--verboseInclude strategy detail and the Evidence Basis used to select the displayed evidence tier.
--jsonReturn the complete machine-readable response for agent and shell integrations.
--no-colorDisable terminal color codes for logs, pipes, and snapshots.
--horizon 5|15|30|60Request one forward outcome horizon instead of the default set.
--limit 1-10Choose how many current confluence leaders to return; only applies to leroy leaders.
How closely does the past match?
Leroy returns the strongest validated comparison available. It never blends exact evidence with broader baselines.
5/5
Match
In this setup, Leroy found the exact same ticker, strategy, market context, and time window 452 times in the archive since 2010.
4/5
Cross-ticker match
The ticker changes, but the strategy pattern and current market context still match. These are comparable setups observed on other symbols.
3/5
Signal pattern
The underlying sensor and strategy pattern match, even though Leroy cannot find a ticker-specific or cross-ticker setup.
2/5
Contextual analogue
The broader context lines up: direction, regime, and strategy conditions match, but the setup itself is not identical.
1/5
Direction baseline
This is the broadest usable comparison: historical outcomes for the same direction, without a close setup match.
If no validated comparison exists, Leroy returns 0/5 — no comparable evidence instead of filling the gap with a guess.
leroy evaluate AAPLBUY evidence
Evidence Strength
AAPL@ $228.87
Exact Setup
452exact observations
Strategy mix
41 evaluatedBuy 25Neutral 10Sell 6
Observed Outcomes
15 Min+1.32%63% win rate · lift +15.80
30 Min+2.87%67% win rate · lift +22.40
60 Min+4.91%71% win rate · lift +31.60
Built for agents, readable by people.
Use the human-readable output in a terminal. Use JSON when another system needs to inspect status, match strength, outcomes, and provenance without parsing presentation text.
$ leroy evaluate ABNB --json
{
"market_memory": {
"matches": [{
"match_strength": {
"score": 5,
"tier": "exact_ticker_setup"
},
"row_count": 50,
"outcomes": { "15": {}, "30": {}, "60": {} }
}]
},
"research_boundary": {
"research_only": true,
"live_order": false
}
}Study the system behind the answer.
The Research Center is where the evidence engine opens up: replay the strategies, compare the context, follow the model, and inspect the paper ledger.
Open Research Center