From 104a5e48290a50705bf929fe2040f3733746dbcb Mon Sep 17 00:00:00 2001 From: Claude Date: Wed, 12 Aug 2026 12:55:27 +0000 Subject: [PATCH] docs(metadata-protocol,objectql): stop hardcoding the shared authoring-rule count in comments MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Three comments asserted that `AUTHORING_RULES` holds 26 shared author-time rules. The registry declares 38, so all three were wrong. Rather than re-hardcode 38 — the same statement that just went stale — the count is dropped and the registry is left as the authority ("all the shared `AUTHORING_RULES`"). The sentences each turn on ALL of the rules being disengaged, never on how many there are, so nothing is lost and the wording cannot expire when rule 39 lands. This matches how the prose docs already refer to the table (content/docs/deployment/validating-metadata.mdx), which names the registry and its file and states no count. Comment-only; no behaviour change. Co-Authored-By: Claude Opus 5 Claude-Session: https://claude.ai/code/session_019Uu3qrB98cMbJ48ae3z9S7 --- packages/metadata-protocol/src/protocol.ts | 4 ++-- packages/objectql/src/plugin.ts | 2 +- 2 files changed, 3 insertions(+), 3 deletions(-) diff --git a/packages/metadata-protocol/src/protocol.ts b/packages/metadata-protocol/src/protocol.ts index b2108ff969..df40555662 100644 --- a/packages/metadata-protocol/src/protocol.ts +++ b/packages/metadata-protocol/src/protocol.ts @@ -2766,7 +2766,7 @@ export type MetadataAuthoringGate = (ctx: MetadataAuthoringGateContext) => void * the CLI's lightweight host-config assembler (`serve.ts`'s auto-register * branch, `new ObjectQLPlugin()` with no options) also leaves `environmentId` * undefined, and it serves an END-USER `PUT /api/v1/meta/*`. Both topologies - * read identically, so the whole #4463 gate — all 26 shared `AUTHORING_RULES` + * read identically, so the whole #4463 gate — all the shared `AUTHORING_RULES` * — was disengaged on a self-hosted app server. #5086 had already moved the * code-only refusal off the same proxy for the same reason ("keying * authorization off a row-scoping key is what made a type-level declaration @@ -3154,7 +3154,7 @@ export class ObjectStackProtocolImplementation implements // with no options) ALSO leaves `environmentId` undefined, and it // serves an end-user `PUT /api/v1/meta/*`. `isHostConfig` → // `shouldBootWithLibrary === false` is the flagship showcase's own - // boot shape, so a self-hosted app server ran all 26 shared + // boot shape, so a self-hosted app server ran all the shared // `AUTHORING_RULES` on exactly nothing — and for a Studio tenant or an // MCP/AI author this gate is not the weakest of four doors, it is the // ONLY one (#4463's filing reason). diff --git a/packages/objectql/src/plugin.ts b/packages/objectql/src/plugin.ts index 90d859619e..b42abd43c3 100644 --- a/packages/objectql/src/plugin.ts +++ b/packages/objectql/src/plugin.ts @@ -73,7 +73,7 @@ export interface ObjectQLPluginOptions { * * **Leave unset on every kernel that serves `PUT /api/v1/meta/*` to end * users.** The default `'environment'` runs the #4463 runtime authoring - * rules (the 26 shared `AUTHORING_RULES` that `os validate` / `os lint` + * rules (the shared `AUTHORING_RULES` that `os validate` / `os lint` * run), which for a Studio tenant or an MCP/AI author is the ONLY * author-time gate — there is no `os lint` for a `sys_metadata` overlay row. *