Skip to main content
The MedLog protocol captures AI use at the level of an inference call. It applies to simple “prompt to model to response” systems as well as multi-stage workflows that use retrieval, tools, rubric-based evaluation, or agentic orchestration. MedLog treats each model invocation as an event and allows events to be linked within the same run or episode. Each MedLog record corresponds to one model invocation and contains the following nine fields.

​
Fields available at invocation

Header

Provenance information, execution context, and system metadata available at inference time.

Model instance

Stable identifiers of the AI model and version, with references to its model card and data sheet.

User identity

The technical process, service, or workflow that invokes the model call.

Target identity

A reference to the entity about which the model produces output.

Inputs

The input data provided to the model.

​
Fields appended later

Internal artifacts

Artifacts generated during inference for technical researchers and MLOps engineers.

Patient- or clinician-facing outputs

The outputs intended for human users.

Outcomes

Records of clinical actions or patient outcomes linked to the model recommendation.

User feedback

Any feedback provided by users, whether structured ratings or free-text comments.
This documentation describes the conceptual nine-field model from the MedLog protocol. The API reference defines the concrete event payloads a collector accepts. See the annotated example record for how the two relate.