The open decision system.

Compile your intent once with an LLM. Then decide on every record forever — classification, scoring, routing — as type-safe JSON with probabilities and confidence, on your own hardware, at zero marginal cost.

Compile once

A large generative LLM scaffolds your execution paths a single time, from a sample of your data, a goal, and a registry of actions. A self-correction loop validates every response — syntax checks, output contracts, import blocklist — before anything is emitted.

Ship one file

The output is a single standalone Python file. Your production environment never imports strot — no framework, no lock-in, just code you can read and version.

Route deterministically

At runtime, a small fast NLU model routes each record to the right action in ~60ms on CPU. Below a 0.40 confidence threshold it raises a FallbackException instead of executing blindly.

import strot

tickets = [
    {"ticket_id": 101, "channel": "email", "body": "I was double charged"},
    {"ticket_id": 102, "channel": "chat",  "body": "App crashes on export"},
]

result = strot.run(
    tickets,
    query="Triage incoming support tickets into the correct workflow.",
    action_registry={
        "billing_dispute":  "Refunds, invoice corrections, double charges",
        "technical_defect": "Crashes, bugs, broken behavior",
    },
    llm="anthropic/claude-opus-4-8",   # used once, on first compile
)

for o in result.outcomes:
    print(o["action"], round(o["confidence"], 2))
# billing_dispute 0.97
# technical_defect 0.94

Benchmarks & research

Honest, reproducible benchmarks — published as we run STROT on new data over time. Wins and losses included.