Skip to main content
QWED v10.0.0 is the current release. Routes stop reporting VERIFIED, or admitting a result, on evidence that does not prove the claim: RAG and process checks are UNVERIFIABLE, a consensus needs two distinct engines, a logic verdict is not admitted, and /verify/math without = is UNVERIFIABLE. The claude_opus provider is removed, and the Go module path ends in /v10. Read the 10.0.0 upgrade steps before you upgrade.

What is QWED?

QWED (Query With Evidence & Determinism) is a trust boundary for AI systems:
  • LLMs can translate user intent into structured claims.
  • QWED verifies those claims with deterministic engines before execution or response.
  • You get proof-backed outcomes instead of probability-only confidence.
QWED is designed for LLM verification, AI agent security, verified tool calls, prompt injection defense, and deterministic transaction verification in high-stakes workflows.
“Do not trust generated output. Verify it.”

Explore common verification problems

LLM verification

Learn how formal verification for LLMs differs from prompting, RAG, and output formatting.

AI agent security

Add pre-execution checks, policy enforcement, and budget controls for agent actions.

Prompt injection defense

Harden your stack against prompt injection, exfiltration, and unsafe execution paths.

QWED MCP

Give MCP clients access to QWED’s deterministic verification libraries.

When to use QWED first

Math and logic

Verify equations, constraints, and logical claims before they reach users.

Code and SQL

Catch unsafe patterns, injection risks, and structural errors before execution.

Agent tool calls

Inspect actions and payloads before external systems are touched.

High-stakes workflows

Add deterministic checkpoints for finance, legal, tax, and regulated systems.

Quick start (5 minutes)

The examples below use the qwed CLI that this installs. The qwedai/qwed-verification:10.0.0 Docker image runs the API server instead; see Self-hosting.
1

Install and configure

Follow Installation and set your provider with LLM configuration.
2

Run first verification

Use Quick start to validate math, logic, code, and SQL.
3

Integrate into production flow

Verification engines at a glance

Math

SymPy-based symbolic verification.

Logic

Z3 SAT/SMT verification with models.

Code

AST and symbolic checks for risky behavior.

SQL

Parser-backed SQL safety and validation.

Schema

Type and shape validation for structured outputs.

Taint

Data-flow tracking for untrusted inputs.

See all engines

Explore core, analysis, and specialized engines.

What’s new in v10

Evidence is not proof

From v10.0.0, a route reports VERIFIED, or admits a result, only on evidence that proves the claim. /verify/rag, /verify/process and the GitHub Action’s verify-process mode are UNVERIFIABLE. /verify/consensus is BLOCKED unless two distinct engines verify the same answer. /verify/logic carries an admission, and it is BLOCKED, because nothing checks that the constraints a model translated mean the question. /verify/image decides a size claim only when the claim states the size, and /verify/math without = is UNVERIFIABLE. Read the 10.0.0 notes →
From v10.0.0, ACTIVE_PROVIDER=claude_opus is refused: set ACTIVE_PROVIDER=anthropic and move its settings to the ANTHROPIC_ ones. A client that acted on status == "VERIFIED" from the routes above now sees UNVERIFIABLE or BLOCKED, the LangChain and CrewAI tools word undecided results differently, and the Go module path moves to /v10. Read the upgrade steps →
  1. Core concepts
  2. Architecture overview
  3. SDKs overview
  4. API overview
  5. Integration guide