Classify every spec as V1 or V2 before launch day

[JarbasAl]

Per VERSIONING.md (see the policy PR): V1 = formalization compatible with the V0 status quo (even if degraded); V2 = not backwards compatible. Every spec file and every open spec PR needs the classification added to its header. Candidate quick reads: parallel-namespace specs behind legacy_namespace gating → V1 with gated V2 sections; renames like fallback.ping/pong → V2 unless dual-emitted.

[JarbasAl]

Draft classification table from PR #61. Evidence was gathered by grepping current ovos-core, ovos-workshop, ovos-dinkum-listener, and plugin checkouts for each spec's normative message types. See PR #61 (CLASSIFICATION.md) for full justifications and the needs-review open questions.

Spec headers will be stamped in each spec's own PR once this classification is reviewed — editing them here would conflict with the 13 in-flight PRs.

On-dev-branch specs

Spec	Classification	Confidence
OVOS-MSG-1	V1	certain
OVOS-SESSION-1	V1	certain
OVOS-SESSION-2	V1	certain
OVOS-PIPELINE-1	V2	certain
OVOS-INTENT-1	V1	certain
OVOS-INTENT-2	V1	certain
OVOS-INTENT-3	V1	certain
OVOS-INTENT-4	V2	certain
OVOS-AUDIO-IN-1	V2	certain
OVOS-STOP-1	V2	certain
OVOS-BRIDGE-1 (on-dev)	needs-review	needs-review

In-flight PR specs

Spec (PR)	Classification	Confidence
OVOS-BRIDGE-1 (PR #43 branch)	V2	needs-review
OVOS-COMMON-QUERY-1 (PR #40)	V2	certain
OVOS-FALLBACK-1 (PR #39)	V2	certain
OVOS-AUDIO-1 (PR #38)	V2	certain
OVOS-PERSONA-1 (PR #37)	V2	certain
OVOS-CONVERSE-1 (PR #25)	V2	certain
OVOS-TRANSFORM-1 (PR #20)	V1	certain
OVOS-CONTEXT-1 (PR #18)	V2	certain
OVOS-INTENT-2 v3 (PR #4)	V1	certain
OVOS-INTENT-4 §11 (PR #45)	V1 section within V2 spec	certain
OVOS-USER-ID-1 (PR #54)	V1	certain

Needs-review open questions

1. OVOS-BRIDGE-1: on-dev version vs PR BRIDGE-1: bus bridge and opaque relay specification #43 branch are significantly different — which version gets the stamp?
2. PIPELINE-1 / AUDIO-1 / STOP-1 / FALLBACK-1: will legacy_namespace-gated dual-emit shims be added (as mentioned in VERSIONING.md re workshop#415 / dinkum#219)? If yes, these become V1+gated-V2 rather than pure V2.
3. OVOS-AUDIO-1: speak → ovos.utterance.speak rename — is a compat shim planned?
4. OVOS-CONVERSE-1: skill.converse.get_response.enable/.disable → session.response_mode — is a migration shim planned?
5. OVOS-FALLBACK-1: does workshop#415/dinkum#219 cover fallback registration topics, or only the skill-side ping/pong?
6. OVOS-STOP-1: mycroft.stop compat — parallel broadcast under a configuration gate planned?

[JarbasAl]

🤖 Auto-generated by Claude Fable 5.1 (claude-fable-5-1) via Claude Code — NOT human-reviewed. Verify before acting. Read from every spec header at dev d9169ee.

Every specification in this repository carries a Version compatibility class in its header, so the ask of this issue is met.

Class	Specs
1	OVOS-INTENT-3, OVOS-SESSION-1, OVOS-SESSION-2, OVOS-GUI-1, OVOS-SCHEDULER-1
2	OVOS-INTENT-1, OVOS-INTENT-2, OVOS-INTENT-4, OVOS-MSG-1, OVOS-BRIDGE-1, OVOS-PIPELINE-1, OVOS-TRANSFORM-1, OVOS-CONTEXT-1, OVOS-CONVERSE-1, OVOS-STOP-1, OVOS-PERSONA-1, OVOS-FALLBACK-1, OVOS-COMMON-QUERY-1, OVOS-AUDIO-IN-1, OVOS-AUDIO-1, OVOS-OCP-1

The draft table above classified OVOS-MSG-1 as V1; the merged header says 2, set by the colon-versus-dotted topic rule (#119), which the migration map in appendix/divergences.md §5.7 records. The ten JarbasHiveMind/architecture specs all carry class 1. Open spec PRs keep the class of the document they amend; a class change is a new spec version under VERSIONING.md, and none of the open PRs makes one.