WF17 Pass-1 watcher/reactor chain still auto-applies ungoverned (same F-a class as the closed #53)

[MSD21091969]

HG URN: urn:moos:session:sam.kernel-proper

Follow-up flagged in #56: PR #56 governs the M6 t_hook event pathway (staging parity with the TIME sweep), but the WF17 Pass-1 watcher→reactor chain (runReactive Kind=="reactor" branch) keeps immediate apply via applyReactiveLocked — skipping §M11/§M12, M8 gates, and any staging. Same F-a class, explicitly out of the t250-Q4 ruling's scope ("react_template fast path").

Work sketch (mirrors #53):

1. Inventory live watcher/reactor chains fleet-wide (probably near-zero, like react_templates were).
2. Either stage reactor proposals through the same stageHookProposalPair-class machinery (watcher nodes have no firing_state — needs an idempotency design decision), or declare+guard the immediate path if a live consumer is load-bearing (the TestRuntime_ReactiveChain KI status flow is test-only).
3. Sam ruling wanted on whether WF17 chains deserve the same parity or an explicit declared fast path.

Wolfram lane, Todo. Refs: #53, #56, fp-fruits finding v.

authored-by: agent:claude-cowork.hp-z440 / session:sam.z440-cowork-workspace / t250-m6-event-governance

[MSD21091969]

CLOSED BY SAM'S RULING (t250 AskUserQuestion, follow-up to t250-Q4): WF17 watcher/reactor chains are DECLARED INTENDED — the explicit LINK wiring IS the declaration.

Evidence that shaped the ruling: the chains are NOT dormant — z440-primary runs 2 watchers + 2 reactors + 1 guard (watcher:tool.verify_baseline, watcher:raw-ki-claim-extract → reactor:emit-claim-extract-task), hp-laptop 1+1. Governing them blind would break live claim-extract machinery.

Doctrine distinction now on record: WF17 chains = deliberately-authored reactive topology (watcher/guard/reactor nodes + explicit triggers/guarded-by LINKs — authoring the topology is the governance act; guards remain the per-chain gate). t_hooks = implicit property-based matching on the owner — those are now staged via #56. The fail-closed unknown-kind routing from #56 keeps any FUTURE firing kind from silently taking the immediate path.

If a WF17 chain ever needs staging semantics, that's a new issue with this ruling as its starting point.

authored-by: agent:claude-cowork.hp-z440 / session:sam.z440-cowork-workspace / t250-resolution-drive