EYE
The original Euler Yet another proof Engine: a mature N3 and RDF Surfaces reasoner with proof output.
EYE Reasoners
EYE has grown from one pioneering Prolog engine into a family of open reasoners for Notation3, RDF, SPARQL 1.2 RL, and Prolog-style rule languages, where an answer can arrive with a proof that another program can check.
The EYE family offers several Notation3 implementations and runtimes. Choose the one that best matches your environment, integration needs, and reasoning workload.
The original Euler Yet another proof Engine: a mature N3 and RDF Surfaces reasoner with proof output.
The original EYE engine compiled to WebAssembly and packaged for browser and Node.js applications.
A compact N3 reasoner for Node.js, browsers, RDF pipelines, streaming messages, and accountable proofs.
A Python port of Eyeling for N3 reasoning from Python code, notebooks, RDFLib graphs, or the command line.
A Rust N3 reasoner that derives conclusions, writes N3 proofs, and checks them, as a CLI, library, or browser app.
A fast, dependency-free forward-chaining N3 reasoner descended from Eyeling and distributed as a static binary.
A newer generation of EYE engines works with Prolog-style facts and rules or with SPARQL. They share one bargain: an answer can come with the reasoning behind it, and a separate checker can verify that proof against the program.
Turns portable ISO Prolog programs into answers and inspectable proofs, with automatic tabling and RDF 1.2 interchange. Documented in The Art of EyeProlog.
A symbolic logic programming language in a subset of Prolog, with query explanations, executable checks, and independently checked proof certificates.
Reasoning you can see: a dependency-free Prolog rule language that combines forward :+ and backward :- rules with checkable proofs.
A standalone, dependency-free Python rule language with forward and backward reasoning, checkable proofs, and a conformance suite.
A small relational reasoner: a Prolog subset with tabled recursion and independently checked proof certificates.
An implementation of SPARQL 1.2 RL (SRL) tracking the W3C draft, with stratification, forward and backward evaluation, and a conformance harness.
These hosted playgrounds let you edit and run examples without installing a local reasoner.
Run the original EYE reasoner in a browser.
Experiment with the compact JavaScript engine.
Reason with the Rust implementation through WebAssembly.
Run SRL rule sets over RDF 1.2 data in the browser.
Run portable Prolog programs with inspectable proofs.
Ask, explain with why, and check assertions.
See derivations and verify their proof certificates.
Write Python rules and run the conformance suite in the browser.
Try tabled recursion and checked proof certificates.
Evaluate Dutch-law questions with Answer, Reason, and Check.
Almost every engine can begin with the same familiar Socrates example: facts and a rule go in; a conclusion comes out, together with the reasoning that supports it.
human(socrates).
mortal(Person) :- human(Person).
?- mortal(Who).
?- why.
@prefix rdfs: <http://www.w3.org/2000/01/rdf-schema#>.
@prefix : <http://example.org/socrates#>.
:Socrates a :Human.
:Human rdfs:subClassOf :Mortal.
{ ?S a ?A. ?A rdfs:subClassOf ?B }
=>
{ ?S a ?B }.
These close neighbours build on the reasoners above: drivers, applications, interchange tools, and test suites.
:+ bottom-up rulesrdf/4 factsLooking for the broader reasoner landscape? Jos De Roo maintains an overview spanning N3, RDF Surfaces, Datalog, SHACL Rules, RIF, and first-order logic: view the full reasoners list.