tok
An alternative concrete syntax for the BPMN 2.0.2 and DMN 1.5 metamodels that expresses process and decision models in plain text. It supports strict validation, compilation to XML, and provision of services to language models via MCP.
Timeline (last 90 days)
tok is an alternative concrete syntax for BPMN and DMN metamodels, expressing BPMN 2.0.2 process models and DMN 1.5 decision models in plain text.
tok documents can be standard-validated through a single engine, compiled to XML with automatic diagrams, and support the MCP protocol.
On a dataset containing 893 BPMN XML files, tok models require 2.2 times fewer tokens than the smallest of the XML files and 7.7 times fewer than the original exported form of the files.
The tok checker achieved a detection rate of 98.2% on 1,411 injected faults, while the combination of XML Schema with bpmnlint achieved a detection rate of 85.8%.
Four language models that had never previously encountered tok, relying only on a one-page reference document, achieved valid tok generation rates of 85.7% to 100% on 87 process descriptions, while the corresponding BPMN XML could be generated for every description.
In the experiment of writing tok from only a one-page reference document, the strongest model used 25% and 19% fewer tokens to write tok than to write BPMN XML.
When the models were configured as agents with a streamlined skill and a checker tool, the valid tok generation rate rose to 97.6% to 100%, and the token savings for the strongest model were 21% and 14%.
If the skill is designed so that the model reads the reference document through a tool, the token overhead is 1.7 to 2.8 times that of the streamlined skill.
Of 47 event logs simulated from public models, 44 can be fitted back to their source models, and tok can import every model discovered from these logs.
Validation of tok is currently concentrated on the two datasets built by the paper itself and on public models, and its performance in large enterprise-level process libraries as well as the willingness of the multi-tool ecosystem to adopt it remain to be seen.
All Facts (10)
tok is an alternative concrete syntax for BPMN and DMN metamodels, expressing BPMN 2.0.2 process models and DMN 1.5 decision models in plain text.
50%Unverifiedtok documents can be standard-validated through a single engine, compiled to XML with automatic diagrams, and support the MCP protocol.
50%UnverifiedOn a dataset containing 893 BPMN XML files, tok models require 2.2 times fewer tokens than the smallest of the XML files and 7.7 times fewer than the original exported form of the files.
50%UnverifiedThe tok checker achieved a detection rate of 98.2% on 1,411 injected faults, while the combination of XML Schema with bpmnlint achieved a detection rate of 85.8%.
50%UnverifiedFour language models that had never previously encountered tok, relying only on a one-page reference document, achieved valid tok generation rates of 85.7% to 100% on 87 process descriptions, while the corresponding BPMN XML could be generated for every description.
50%UnverifiedIn the experiment of writing tok from only a one-page reference document, the strongest model used 25% and 19% fewer tokens to write tok than to write BPMN XML.
50%UnverifiedWhen the models were configured as agents with a streamlined skill and a checker tool, the valid tok generation rate rose to 97.6% to 100%, and the token savings for the strongest model were 21% and 14%.
50%UnverifiedIf the skill is designed so that the model reads the reference document through a tool, the token overhead is 1.7 to 2.8 times that of the streamlined skill.
50%UnverifiedOf 47 event logs simulated from public models, 44 can be fitted back to their source models, and tok can import every model discovered from these logs.
50%UnverifiedValidation of tok is currently concentrated on the two datasets built by the paper itself and on public models, and its performance in large enterprise-level process libraries as well as the willingness of the multi-tool ecosystem to adopt it remain to be seen.
50%