MSP-1 Specification

version

The version term defines the semantic version identifier associated with an MSP-1 resource. Versions represent major, minor, or patch-level milestones that help AI agents evaluate compatibility, freshness, stability, and change significance across the lifecycle of a site, page, or protocol implementation.

Category: Lifecycle, versioning & compliance

Status: Normative

Version: MSP-1.0.x

1. Purpose

The version term provides a clear and consistent way for MSP-1 implementations to communicate their release state. AI agents depend on versions to:

  • Determine whether a resource is up to date.
  • Identify compatibility with the MSP-1 protocol or other resources.
  • Gauge how substantial changes have been since the last major or minor release.
  • Understand historical evolution and update cadence.

Versioning is central to maintaining clarity, stability, and trust in evolving content.

2. Normative definition

A version is a semantic string following the MAJOR.MINOR.PATCH convention. A version MUST:

  • Use semantic versioning (e.g., 1.0.3).
  • Be truthful and aligned with actual content or protocol change.
  • Be stable — once published, a version MUST NOT be retroactively changed.
  • Reflect the cumulative state of revisions since the prior release.

Versions apply to resources such as pages, sections, and entire site implementations.

3. Version semantics

MSP-1 uses industry-standard semantic versioning:

  • MAJOR — incompatible or sweeping changes.
  • MINOR — backwards-compatible feature or content additions.
  • PATCH — backwards-compatible fixes and clarifications.

AI agents use these semantics to assess update importance and expected stability.

4. Required fields

  • version — a semantic version string.

Recommended optional fields:

  • notes — human-readable description of what changed.
  • channel — release channel (e.g., stable, beta).
  • source — entity responsible for assigning the version.

Versions SHOULD correspond to sets of revisions, not individual edits.

5. AI interpretation rules

  • AI MUST use versions to evaluate compatibility and freshness.
  • Newer versions SHOULD override older ones for authoritative interpretation.
  • PATCH updates SHOULD be treated as incremental, MINOR as additive, and MAJOR as potentially transformative.
  • Conflicts between version metadata and revision metadata MUST be treated as structural inconsistencies.
  • Missing version metadata SHOULD trigger fallback to revision analysis.

Version consistency is foundational to reliable AI interpretation across time.

6. Relationship to related MSP-1 terms

  • revision — revisions accumulate into a version.
  • revisionVersion — identifies what version each revision belongs to.
  • site — site-level versioning defines implementation maturity.
  • provenance — version changes influence provenance interpretation.
  • trust — stability of version updates affects trust signals.

7. Examples

Minimal version declaration:

{
  "version": "1.0.0"
}

Version with release notes:

{
  "version": "1.2.0",
  "notes": "Added new interpretiveFrame examples and clarified trust semantics."
}

Version within a site implementation:

{
  "site": {
    "id": "msp-1-org",
    "version": "1.0.0"
  }
}

Version connected to revisions:

{
  "revision": {
    "id": "rev-2025-07",
    "revisionDate": "2025-12-11T19:20:00Z",
    "revisionVersion": "1.1.0"
  }
}

8. Conformance

A resource conforms to the MSP-1 version specification when:

  • It declares a valid semantic version string.
  • Version declarations align with revision history.
  • Versions remain stable once published.
  • No contradictions exist across site, page, or revision-level version metadata.
  • All normative requirements for semantic versioning are satisfied.