Kairos

Kairos is an open-source real-time fraud and risk decisioning engine that evaluates transactions and behaviors.

Kairos is an open-source fraud and risk decisioning platform that evaluates sign-ups, logins, orders, payments, and payouts in real time. It returns actions like approve, step up, review, or block along with risk scores and detailed reasons. The platform combines Grayson, a decision model that answers plain-language questions and handles unstructured data, with traditional machine learning models like logistic regression and XGBoost-compatible gradient-boosted trees. Coding agents connect via Model Context Protocol to build workflows, propose model experiments, test them on historical data, and ship winners. Developers integrate Kairos using TypeScript and Python SDKs, an HTTP API, and a CLI, with self-hosting options available via Docker.

Kairos at a glance

Best for: Software engineer
Runs on: API, CLI, self-hosted, open source
Pricing: Free
Released: November 2026
Category: AI coding tool
Website: github.com/finic-ai/kairos

Key features

Real-time fraud and risk decisions
Agent-driven workflow and experiment tuning
Grayson decision model for unstructured data
Traditional ML models like logistic regression and XGBoost
Queues and cases for manual review
TypeScript and Python SDKs
HTTP API and CLI
Self-hosting via Docker

Use it for

Deciding sign-ups, logins, orders, and payments in real time
Reviewing risky orders and learning from chargebacks
Building and tuning decision workflows with an AI agent
Scraping insights from unstructured customer data

Alternatives to Kairos

Project STROT — Runs locally at zero marginal cost with typed JSON outputs and confidence-based abstention.
Kilo Code — Open-source agent with any model and no markup on inference
Kiro — Spec-driven coding with property-based tests to catch hidden bugs
PullRequest — Combines AI risk analysis with vetted human experts to catch critical vulnerabilities.
AiSOC — Runs locally without API keys with bundled models and auditable reasoning ledgers.
Kora Doctor — Surfaces waste across tool usage, context growth, and repeated inference without an account.
HackerOne — Combines AI agents with expert human researchers to prioritize and fix real risks.
Genkit — Provides unified APIs and local developer tools for AI application workflows.
AI Crimes — Practice catching real-world AI coding mistakes on anonymized snippets.
Farol — Ranks risky files by reach and findings to guide code reviews.
Giskard — It offers automated vulnerability scanning and behavioral testing for LLM agents.
ADRs4AI — Captures and version-controls the reasoning behind human-AI collaboration.

All Kairos alternatives