From 693e25ef05113a0667e5c4ce44faafe92dbe47fc Mon Sep 17 00:00:00 2001 From: Mark Beacom Date: Mon, 10 Aug 2026 11:56:59 -0400 Subject: [PATCH 1/3] docs: sync to v0.5.0 and carry the cutover's evidence forward MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit The site redeployed from main seconds after the release merge, so adrkit.dev has been advertising v0.4.0 while npm serves 0.5.0. This closes that, and the same staleness in the README, RELEASING, and DISTRIBUTION. Two entries needed rewriting rather than a version bump. The README still described markers as reaching `adr explain` "and nothing else", which ADR-0022 reversed, and CLAUDE.md still headed the section "check/CI extension proposed" — both now describe the shipped v0.5.0 scope, name ADR-0023's format rule, and credit @aballiet and @davesheffer for the three pull requests behind it. `ci.mdx` named the commit `v0` peels to, which moved with the tag, and DISTRIBUTION's P1/P2 prerequisites are re-stated against the published 0.5.0 package rather than 0.4.0 — both verified against npm and the tag, not assumed. RELEASING gains two notes from this cutover: that #104's dry-run failure was confirmed dry-run-only by comparing the packed shasum to the registry's before tagging, and that the MCP registry has no step in release.yml, so `dev.adrkit/mcp` keeps serving the previous version until a human publishes. DISTRIBUTION's §A submission record is deliberately untouched. It states a verified outcome with a date, so it is written after the registry publish, not before. Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com> Signed-off-by: Mark Beacom --- CLAUDE.md | 14 +++++++------- README.md | 25 ++++++++++++++++--------- docs/DISTRIBUTION.md | 10 +++++----- docs/RELEASING.md | 23 ++++++++++++++++++++--- site/src/components/Hero.astro | 2 +- site/src/content/docs/ci.mdx | 4 ++-- site/src/content/docs/index.mdx | 2 +- site/src/content/docs/quickstart.mdx | 2 +- 8 files changed, 53 insertions(+), 29 deletions(-) diff --git a/CLAUDE.md b/CLAUDE.md index 3802adef..5ab4422a 100644 --- a/CLAUDE.md +++ b/CLAUDE.md @@ -2,7 +2,7 @@ Decision memory for human- and agent-authored plans — machine-readable ADRs that are enforceable in CI and legible to agents, without leaving git. -Status: early — phases 0–6 landed and v0.4.0 is public. `@adrkit/core`, +Status: early — phases 0–6 landed and v0.5.0 is public. `@adrkit/core`, `@adrkit/evaluator`, `@adrkit/cli` (`lint`, `new`, `graph`, `explain`, `check`, `queue`, `migrate --from madr`, `evaluate`) are published on npm, as is the independently versioned `@adrkit/spec-kit` Spec Kit extension (0.1.2); the @@ -17,16 +17,16 @@ driven through the official MCP Inspector. The Inspector defaults to the 2025 era; select the modern one with `"protocolEra": "modern"` (or `"auto"`) in the server's entry in the Inspector's `mcp.json` — there is no CLI flag for it. -## Inbound `@adr` markers (v0.4.0 explain; check/CI extension proposed) +## Inbound `@adr` markers (v0.5.0: explain, check, and CI) A file can declare the decision it lives under by putting `@adr 0012` on a dedicated comment line inside its first 8192 bytes. v0.4.0 shipped that inbound -edge for `adr explain ` under -[ADR-0021](./docs/adr/0021-resolve-inbound-source-annotations-without-changing-the-schema.md) -and [#97](https://github.com/mbeacom/adrkit/pull/97). This branch proposes +edge for `adr explain ` under ADR-0021 and +[#97](https://github.com/mbeacom/adrkit/pull/97). v0.5.0 extended the same +resolution to `adr check` and the governing-decisions Action under [ADR-0022](./docs/adr/0022-scan-inbound-markers-in-check-and-ci-without-giving-them-exit-code-authority.md), -which extends the same resolution to `adr check` and the governing-decisions -Action. No schema change: `AdrFrontmatter`, +which **supersedes ADR-0021** — read 0022, not 0021, for the current scope. No +schema change: `AdrFrontmatter`, `AffectsType`, and `schema/adr.schema.json` are untouched. Two properties are load-bearing and easy to break: diff --git a/README.md b/README.md index ab78fd39..f68a59d2 100644 --- a/README.md +++ b/README.md @@ -225,24 +225,31 @@ different artifact from a heading convention. That is the whole thesis. Early, under active development, and deliberately honest about what is proven. -- **Published — v0.4.0 on npm.** The schema, `@adrkit/core`, `@adrkit/cli`, +- **Published — v0.5.0 on npm.** The schema, `@adrkit/core`, `@adrkit/cli`, the deterministic Pass 0 `@adrkit/evaluator`, and the read-only `@adrkit/mcp` server are all implemented and released. The MCP server speaks both protocol eras and passed real-session dogfood against the published artifact on each, driven through the official MCP Inspector ([ADR-0018](docs/adr/0018-adopt-mcp-sdk-v2-and-serve-protocol-revision-2026-07-28-dual-era.md)). -- **New in v0.4.0, and at rung 1 only.** A file can declare the decision it +- **New in v0.5.0, and at rung 1 only.** A file can declare the decision it lives under with an `@adr ` marker on a dedicated comment line, and - `adr explain` resolves that inbound edge alongside the `affects` patterns that - already matched the path - ([ADR-0021](docs/adr/0021-resolve-inbound-source-annotations-without-changing-the-schema.md)). - The feature reaches `adr explain` and nothing else: `adr check`, the Action, - and the Spec Kit context script do not scan markers, so no CI semantics move. + `adr explain`, `adr check`, and the governing-decisions Action all resolve that + inbound edge alongside the `affects` patterns that already matched the path + ([ADR-0022](docs/adr/0022-scan-inbound-markers-in-check-and-ci-without-giving-them-exit-code-authority.md), + superseding [ADR-0021](docs/adr/0021-resolve-inbound-source-annotations-without-changing-the-schema.md)). + A marker is read only where the file's own format hides it — never inside a + fenced block, and in markdown only from `