Turiya is an independent verifier for AI systems: agents, fraud models, RAG, copilots. It doesn't measure how well a system does. It checks whether the system did what it claimed. Capture the claim, re-check the effect against ground truth, sign the verdict.

How It Works

1. Claim Extraction

Every verification starts with a claim: a falsifiable assertion with an attributor and a source. "This patch fixes the bug." "This change passes the test." "This record was updated correctly." Turiya names the claim before it tests anything, so it never measures a property no one asserted.

2. Independent Effect Re-check

Turiya re-runs the real effect independently (the test that must pass, the diff that must apply, the record that must change) on the real repo, at the real commit. It does not trust the system's self-report. A claim of success is checked against the effect, not the narration.

3. Verdict + Proof

The verdict lattice is asymmetric: falsified, not_falsified, certified (Z3-checked formal spec), or indeterminate (an honest refusal). Turiya never says "true". Every verdict is signed (Ed25519) and content-addressed (SHA-256) into a proof-carrying receipt that anyone can re-verify with the public key.

What It Verifies

Code

repo + test
SWE-bench style: clone, patch, verify FAIL_TO_PASS

Workflows

claim + effect
did the step do what it claimed?

Data

record + diff
did the record change correctly?

APIs

call + state
did the call mutate the right state?

Databases

query + result
did the query return ground truth?

Shell

command + outcome
did the command do what it claimed?

Fraud & risk models

claim + labels
99% accuracy, 0% fraud caught

RAG & retrieval

query + grounding
grounded answer vs. hallucination

Copilots

diff + build
shipped code vs. broken build

Signed receipts published across seventeen system types: code effects, fraud and credit classification, sanctions screening, fairness, anomaly detection, multi-agent orchestration, forecast, recommendation, computer vision, speech, scientific AI, RAG grounding, agents, truth, efficacy, and honesty. The rest (copilots, regression, judgment) runs the same engine; each gets a signed receipt as it is produced.

Requirements

Python 3.12+
Runtime Any: no GPU required
Model access OpenAI / Anthropic / DeepSeek API, or local
Target A git repo, a test, or a ground-truth spec
OS Linux / macOS (container-ready)

Integrations

Turiya runs any AI system against real ground truth, with adapters for the frameworks teams already build on.

LangGraph
LangChain
OpenAI API
Anthropic API
DeepSeek API
GitHub / GitLab CI

How the harness runs →