docs(vision): clear post-F1 staleness items (DRAFT header, §2/§3/§4 forward-tense, PR-N placeholders)

[AdaWorldAPI]

Summary

Closes the staleness debt Sprint C agent flagged but didn't fix (out of its §7-only scope) in PR #311's report:

Other stale items found (not fixed, per instructions):

• DRAFT — pending review (2026-04-28) line 1 and the §2 anchor as of 2026-04-28: doc still labels itself a draft from before F1 shipped.
• §2 latency row still says F1 numbers as if forthcoming.
• §3 phase descriptions still reference placeholder gates PR-1, PR-2, PR-4 instead of real lance-graph PR numbers.
• §4 still says "the benchmark harness lands as part of F1" in the future tense.

All five fixed in this PR. Pure docs change. Diff: +41 / −26 in one file.

What changed (every edit cites a concrete PR or file path; tone rule preserved)

Where	Before	After
Header	DRAFT — pending review (2026-04-28)	Status: F1 parity shipped 2026-04-30. F1 latency benchmark not yet started. F2 is a posture.
§2 anchor	as of 2026-04-28	as of 2026-04-30 (post-F1 parity ship)
§2 latency cell	Designed to match C# direct-MySQL; F1 numbers	Designed to match C# direct-MySQL; benchmark pending
§2 caveat	F1 publishes the first numbers. Until then, do not quote latency claims.	Disambiguates the two F1 sub-deliverables: parity (shipped) vs latency benchmark (not started).
§3 F1	Forward-tense narrative	Shipped 2026-04-30. Cross-link to §7's as-shipped architecture.
§3 F2	gated upstream by lance-graph PR-1 / PR-2	Real PR numbers (#278 + #280 + #284 for RLS, #278 + #302 for audit) plus a Status today line stating: lance-graph in production; medcare-rs adopter not yet open; F2 is a posture, not a delivery.
§3 F3	gated upstream by lance-graph PR-4	lance-graph PR #278 + #280 for the parser + hardening. Status today clarifies parser stub is on main; medcare-rs adopter is future.
§4	The benchmark harness lands as part of F1. (forward)	F1 parity (correctness) shipped 2026-04-30; F1 latency benchmarking on the fixed corpus has not been started. The two are separately-scoped F1 sub-deliverables.
§4 footnote	until F1 numbers are published	until the F1 latency benchmark report is published (more specific)
§4 next list	What we will measure in F1	What we will measure when the F1 latency benchmark runs

What this PR does NOT touch

• §4, §5 (risks), §6 (NOT promising) — no claims expired. §5 / §6 are time-invariant content.
• §7 (next deliverable) — landed clean in PR docs(vision): mark F1 shipped, restate next deliverable as F2 RBAC wiring #311 already.
• Performance numbers — none claimed. The §4 "do not quote unbenchmarked numbers" rule is preserved verbatim.
• F4, F5 phase descriptions — genuinely future, no claims to update.
• Tone — every change cites a concrete PR number or file path. No marketing language introduced. The doc's "Tone: brutally honest. No hype. No marketing claims that have not been benchmarked." holds.

Verification

Markdown renders cleanly (table column-pipe alignment retained; cell width drift is normal for content edits).

Files changed

• .claude/medcare-foundry-vision.md — +41 / −26

Cross-link

• PR docs(vision): mark F1 shipped, restate next deliverable as F2 RBAC wiring #311 — the §7 fix that motivated this cleanup. Sprint C agent's report explicitly listed these as deferred.

Generated by Claude Code

---

Generated by Claude Code