Part of the WebMEM Protocol
Location: /specification/sdt/yaml-in-html/classes/policyfragment/
Last Updated: 2025-08-02
Overview
A PolicyFragment encodes structured policy rules, conditions, exceptions, and logic paths in a retrievable, explainable format. These fragments allow AI systems to evaluate and explain regulations, enforcement criteria, and exception handling with trust-scored provenance and glossary alignment.
Purpose: To publish policy logic and regulatory conditions in a format that supports AI onboarding, user-facing explanation, and machine-readable trust evaluation.
Common Applications:
- Medicare penalties, protections, or eligibility timelines
- Public benefit rules with exception handling
- Insurance regulations and enrollment windows
- Education and financial aid policies
- Local government programs and mandates
Fields Block
| Field | Description |
|---|---|
policy_name |
Canonical name of the policy or rule |
rule |
Default logic or requirement imposed by the policy |
exceptions |
List of exceptions, exemptions, or override conditions |
glossary |
Glossary term alignment for rule logic |
retrieval_hint |
Memory cue for surfacing this policy fragment |
Logic Model (Optional)
logic_model:
if: "no_coverage and months_delayed > 0"
then: "Apply penalty"
else_if: "has_creditable_coverage"
then: "No penalty applied"
else: "Evaluate with exception rules"ProvenanceMeta Block
| Field | Description |
|---|---|
id |
Unique ID for the provenance record |
title |
Human-readable source name |
creator |
Publishing organization |
license |
Usage rights (e.g., Public Domain) |
published |
ISO 8601 date published |
retrieved |
Date this content was accessed |
digest |
Digest ID this fragment belongs to |
entity |
(Optional) Rule subject or policy ID |
fragment_scope |
Typically semantic-digest |
glossary_scope |
Namespace of relevant definitions or terms |
Why PolicyFragments Matter
- Retrievable: Policies can be directly queried and explained
- Logic-Aware: Supports conditional evaluation of rules and exceptions
- Explainable: Enables AI to justify regulatory outcomes to users
- Glossary-Bound: All key terms are defined and scoped
- Federated-Ready: Can power legal, medical, and public policy interfaces