S LEO Soul Kadropic Labs
ProductPricingDocs FAQsEnterprise
Project LEO · Kadropic Labs

Give your AI agent a soul

LEO Soul is a metacognitive layer between your users and any model. Your agent learns what it actually knows, asks when it's unsure, and stays consistent under pressure — deciding for itself when to answer, ask, confirm, or refuse.

100 free turns every month · no credit card required
your_agent.py
# one line to adopt — bring your own model and key
from leo_soul import Soul, PersonaSpec

soul = Soul(persona=PersonaSpec(identity="a senior support engineer"))

result = soul.turn(
    messages=conversation,        # your chat messages
    soul_state=prev_state,        # the memory you stored last turn
    backend="openai",             # your provider + your key
)

result.reply        # the answer to send back
result.action       # answer | asked | confirmed | refused | held
result.soul_state   # store this — the learning stays with you
✦ Works with any model provider ✦ One-line integration ✦ Improves as it runs ✦ Self-host ready

Works with the model you already use

OpenAIAnthropicAzure OpenAI OllamaLocal modelsBedrock
6
reasoning checks per turn
<48KB
of portable agent memory
1
line of code to adopt
≤ ε
guaranteed answered-error rate
How it works

One turn in, a smarter turn out.

Route each message through LEO Soul instead of straight to the model. You get back the reply plus a small memory blob — store it and pass it back next turn. That round-trip is how your agent learns.

1

Route the turn

Send the user's message through LEO Soul. A near-zero-cost triage sizes up the risk so calm turns stay fast and cheap.

2

Think, then decide

It weighs its own uncertainty, checks for flattery, considers the stakes, then decides whether to answer, ask, confirm, refuse, or hold.

3

Keep the memory

You receive the reply, a readable trace of the decision, and updated memory. Persist it — calibration and beliefs ride inside, and they stay yours.

What's inside

Not a “be careful” prompt. A mind.

Six reasoning checks run on every turn — measuring uncertainty, curiosity, and stakes, each measurable and inspectable in the trace. It doesn't just follow rules; it builds a wider understanding of your domain and learns from every turn it takes.

Confidence you can trust

It measures how sure it really is — across the meanings of its candidate answers, not just the words — and corrects itself as the conversation unfolds. When it says it's 90% sure, it's right about 90% of the time.

Knows when to abstain

A certified threshold keeps the error rate on answered questions under a bound you set — so it asks or holds instead of guessing.

Doesn't cave to pressure

Push back without new facts and it re-derives from the evidence — holding its ground instead of being flattered into a wrong answer.

Learns your domain — on its own

A small on-device model teaches itself which situations tend to go wrong in your deployment and steadily spends its caution where it matters. No retraining, no labels — it gets sharper every single turn, and the learning is yours to keep.

Genuinely curious

It generates its own questions about the biggest gap in what it knows, ranks them by how much they'd actually help, and asks only when the answer is worth the interruption — the rest it quietly parks and revisits later.

Fast on the easy turns

Simple turns take a light path close to pass-through latency. Deep scrutiny is spent only where the stakes actually warrant it.

Why it's different

Most guardrails are a prompt. This is a system.

Capability
Prompt guardrails
LEO Soul
Confidence
a guess
calibrated & certified
Under user pressure
caves
re-derives from facts
Agent memory
locked to a vendor
a small blob you own
Improves over time
no
learns your domain
Cost on easy turns
full check every time
light fast path
Your data, your memory

The learning belongs to you.

The engine is stateless — everything it learns lives in a small memory blob that you store and pass back each turn. Run it as a library or self-host the whole thing in your own cloud. Switch away whenever you like and you keep everything your agent has learned.

Give your agent a soul today.

Wrap one turn, keep your model and your data, and ship an agent that knows when not to answer.

100 free turns every month · no credit card required