Compose a specialist agent for every layer of investing.
One agent runs quant signals, another reads filings, another manages risk, another executes trades through your brokerage, another runs exit strategy — every agent loaded with your investment thesis, every gated trade auditable, your humans hold the deal-review gate.
Best fit for · Investors and trading desks running multi-strategy analysis with deal review, risk gates, and brokerage executionvs Perplexity Computer →Five concrete tasks.
- 01
Specialist agents for every layer of investing.
Compose the pack you need: a Quant for algorithmic signals, a Fundamentals analyst for filings + earnings, a Risk Manager for exposure + drawdown, a Trader for brokerage execution, an Exit Strategy specialist for stop-loss and take-profit. Add or drop specialists per strategy — long-only, market-neutral, event-driven all want different shapes.
- 02
Strategy thesis loads into every agent.
A strategy custodian (or your thesis memory) holds the investment thesis, risk limits, position sizing rules, and disqualification criteria. Every other agent — Quant, Fundamentals, Trader, Exit — loads it before acting. Trades that violate the thesis don't get past the gate.
- 03
Multi-strategy signal generation in parallel.
Quant runs algorithmic signals (momentum, mean-reversion, factor models). Fundamentals reads filings + earnings + cash flow. Technical patterns surface from chart analysis. The pack triangulates across strategies before recommending a position — single-strategy bets don't pass the deal-review gate.
- 04
Automated execution with human-gated approvals.
Trader executes through your brokerage API (IBKR, Alpaca, Tradier, Robinhood) once a trade clears the deal-review gate. Position size + stop loss attached at entry. Risk Manager monitors drawdown live; trips a circuit breaker when limits hit.
- 05
Exit strategy that doesn't forget the plan.
Exit Strategy agent watches every open position against the original thesis. Stop-loss, take-profit, trailing stop, time-based exit — defined at entry, executed without sentiment. Performance Analyst attributes P&L back to which signals worked.
Setup once, then watch it run.
Concrete operator setup, the phases the pack moves through, and where you stay in the loop.
Universe + signals
Quant scans the universe against algorithmic signals; Fundamentals reads filings + earnings + cash flow. Market Data feeds both. Triangulated candidates surface to the deal-review queue.
Deal review
Investment Lead + Risk Manager review thesis fit, position size, hedge requirements. Three-tier approval before any trade executes.
Execution
Approved trades execute through your brokerage API. Stop-loss + take-profit attached at entry; Brokerage Sync tracks fills.
Live monitoring
Risk Manager watches every open position against drawdown limits; Exit Strategy fires take-profits and stop-losses without sentiment. Circuit breakers on limit breach.
Performance
Performance Analyst attributes P&L to signals + strategies weekly. Calls out which patterns worked. Proposes thesis refinements.
Your pack, your workflow.
Workflows are markdown that reference the roles in your pack. Below is one example shape - yours can have different agents, different steps, different cadence.
A multi-strategy analyst desk with brokerage execution — composable, thesis-loaded, with risk gates and audit trail on every trade.
Things teams actually ask.
Run this pack on your team's work.
Closed-beta cohorts are small. Tell us about your work and we'll configure the pack for what you actually do.
Request beta access