All-Blockchains
diff --git a/‎.gitattributes‎
Lines changed: 3 additions & 1 deletion b/‎.gitattributes‎
Lines changed: 3 additions & 1 deletion
diff --git a/‎.github/aw/actions-lock.json‎
Lines changed: 34 additions & 0 deletions b/‎.github/aw/actions-lock.json‎
Lines changed: 34 additions & 0 deletions
diff --git a/‎.github/skills/api-diff/SKILL.md‎
Lines changed: 72 additions & 19 deletions b/‎.github/skills/api-diff/SKILL.md‎
Lines changed: 72 additions & 19 deletions
diff --git a/‎.github/skills/editorial-scoring/SKILL.md‎
Lines changed: 109 additions & 0 deletions b/‎.github/skills/editorial-scoring/SKILL.md‎
Lines changed: 109 additions & 0 deletions
diff --git a/‎.github/skills/generate-changes/SKILL.md‎
Lines changed: 129 additions & 0 deletions b/‎.github/skills/generate-changes/SKILL.md‎
Lines changed: 129 additions & 0 deletions
@@ -60,4 +60,6 @@
 #*.pdf   diff=astextplain
 #*.PDF   diff=astextplain
 #*.rtf   diff=astextplain
-#*.RTF   diff=astextplain
+#*.RTF   diff=astextplain
+
+.github/workflows/*.lock.yml linguist-generated=true merge=ours
@@ -0,0 +1,34 @@
+{
+  "entries": {
+    "actions/download-artifact@v8.0.1": {
+      "repo": "actions/download-artifact",
+      "version": "v8.0.1",
+      "sha": "3e5f45b2cfb9172054b4087a40e8e0b5a5461e7c"
+    },
+    "actions/github-script@v8": {
+      "repo": "actions/github-script",
+      "version": "v8",
+      "sha": "ed597411d8f924073f98dfc5c65a23a2325f34cd"
+    },
+    "actions/setup-dotnet@v5.2.0": {
+      "repo": "actions/setup-dotnet",
+      "version": "v5.2.0",
+      "sha": "c2fa09f4bde5ebb9d1777cf28262a3eb3db3ced7"
+    },
+    "actions/upload-artifact@v7": {
+      "repo": "actions/upload-artifact",
+      "version": "v7",
+      "sha": "bbbca2ddaa5d8feaa63e36b76fdaad77386f024f"
+    },
+    "github/gh-aw-actions/setup@v0.65.1": {
+      "repo": "github/gh-aw-actions/setup",
+      "version": "v0.65.1",
+      "sha": "1831bcd4f397da99da6e6e6b65687557a1a37ac7"
+    },
+    "github/gh-aw-actions/setup@v0.65.6": {
+      "repo": "github/gh-aw-actions/setup",
+      "version": "v0.65.6",
+      "sha": "31130b20a8fd3ef263acbe2091267c0aace07e09"
+    }
+  }
+}
@@ -1,30 +1,83 @@
 ---
 name: api-diff
-description: Generate an API comparison report between two .NET versions using the RunApiDiff.ps1 script. Invoke when the user asks to run, create, or generate an API diff.
-disable-model-invocation: true
+description: >
+  Download, inspect, and diff .NET APIs for a given build or release. Use
+  `dotnet-inspect` to verify that APIs actually exist in the shipped packages,
+  detect missed reverts, and generate before/after API diffs. Use
+  `RunApiDiff.ps1` when a full markdown diff report is needed.
 ---
 
-# API Diff Generation
+# API Downloading, Verification, and Diffing
 
-Map the user's request to parameters for `release-notes/RunApiDiff.ps1` and run it. See [release-notes/RunApiDiff.md](../../../release-notes/RunApiDiff.md) for the full parameter reference.
+Use this skill when you need evidence about the **actual public API surface** of a .NET build.
 
-When no versions are mentioned, run with no parameters — the script auto-infers versions.
+This is primarily for:
+
+- confirming whether an API really shipped in the build binaries/ref packs
+- catching reverts or incomplete rollouts before they end up in release notes
+- generating before/after API diffs for a release milestone
+
+## Preferred workflow
+
+### 1. Query the right build first
+
+Do **not** trust the locally installed SDK for preview work. Query the target build's packages directly.
+
+Use the process in [api-verification.md](../release-notes/references/api-verification.md):
+
+1. Generate or read `build-metadata.json`
+2. Get `nuget.source` and the package versions for the target release
+3. Run `dotnet-inspect` against those exact packages
+
+### 2. Verify APIs with `dotnet-inspect`
+
+Typical tasks:
+
+- **Find a type or member** to confirm the API exists
+- **Compare versions** to see what changed between previews or between RC and GA
+- **Validate naming** before writing prose or code samples
+
+Examples:
+
+```bash
+# Find a type in the runtime ref pack
+dnx dotnet-inspect -y -- find "*AnyNewLine*" \
+  --package "Microsoft.NETCore.App.Ref@${VER}" \
+  --source "$FEED"
+
+# Verify members on a type
+dnx dotnet-inspect -y -- member RegexOptions \
+  --package "Microsoft.NETCore.App.Ref@${VER}" \
+  --source "$FEED" \
+  -k field
+
+# Diff public APIs between two versions
+dnx dotnet-inspect -y -- diff \
+  --package "Microsoft.NETCore.App.Ref@11.0.0-preview.2..11.0.0-preview.3" \
+  --source "$FEED"
+```
+
+If the API is missing from the target build, treat that as a serious signal: it may have been renamed, kept internal, or reverted.
+
+### 3. Use `RunApiDiff.ps1` for full diff reports
+
+When the user wants the markdown-ready, repo-shaped API diff output, use `release-notes/RunApiDiff.ps1`. See [release-notes/RunApiDiff.md](../../../release-notes/RunApiDiff.md) for the full parameter reference.
 
 ## Mapping natural language to parameters
 
-| User says | Parameters |
-|---|---|
-| "generate the next API diff" | *(none)* |
-| ".NET 10 GA vs .NET 11 Preview 1" | `-PreviousMajorMinor 10.0 -CurrentMajorMinor 11.0 -CurrentPrereleaseLabel preview.1` |
-| "net9.0-preview6 to net10.0-preview5" | `-PreviousMajorMinor 9.0 -PreviousPrereleaseLabel preview.6 -CurrentMajorMinor 10.0 -CurrentPrereleaseLabel preview.5` |
-| ".NET 10 RC 2 vs .NET 10 GA" | `-PreviousMajorMinor 10.0 -PreviousPrereleaseLabel rc.2 -CurrentMajorMinor 10.0` |
-| "10.0.0-preview.7.25380.108 to 10.0.0-rc.1.25451.107" | `-PreviousVersion "10.0.0-preview.7.25380.108" -CurrentVersion "10.0.0-rc.1.25451.107"` |
-
-- **GA** or no qualifier → omit the PrereleaseLabel parameter
-- **Preview N** / **previewN** → `-PrereleaseLabel preview.N`
-- **RC N** / **rcN** → `-PrereleaseLabel rc.N`
-- **netX.Y-previewN** (TFM format) → `-MajorMinor X.Y -PrereleaseLabel preview.N`
-- Full NuGet version strings → use `-PreviousVersion` / `-CurrentVersion` directly
+| User says                                             | Parameters                                                                                                             |
+| ----------------------------------------------------- | ---------------------------------------------------------------------------------------------------------------------- |
+| "generate the next API diff"                          | _(none)_                                                                                                               |
+| ".NET 10 GA vs .NET 11 Preview 1"                     | `-PreviousMajorMinor 10.0 -CurrentMajorMinor 11.0 -CurrentPrereleaseLabel preview.1`                                   |
+| "net9.0-preview6 to net10.0-preview5"                 | `-PreviousMajorMinor 9.0 -PreviousPrereleaseLabel preview.6 -CurrentMajorMinor 10.0 -CurrentPrereleaseLabel preview.5` |
+| ".NET 10 RC 2 vs .NET 10 GA"                          | `-PreviousMajorMinor 10.0 -PreviousPrereleaseLabel rc.2 -CurrentMajorMinor 10.0`                                       |
+| "10.0.0-preview.7.25380.108 to 10.0.0-rc.1.25451.107" | `-PreviousVersion "10.0.0-preview.7.25380.108" -CurrentVersion "10.0.0-rc.1.25451.107"`                                |
+
+- **GA** or no qualifier -> omit the `PrereleaseLabel` parameter for `RunApiDiff.ps1`; if a downstream api-diff release label is needed, use `ga`
+- **Preview N** / **previewN** -> `-PrereleaseLabel preview.N` (for example, `preview.4`, not `preview4`)
+- **RC N** / **rcN** -> `-PrereleaseLabel rc.N`
+- **netX.Y-previewN** (TFM format) -> `-MajorMinor X.Y -PrereleaseLabel preview.N`
+- Full NuGet version strings -> use `-PreviousVersion` / `-CurrentVersion` directly
 - The "previous" version is always the older version; "current" is the newer one
 
 ## Running the script
@@ -33,4 +86,4 @@ When no versions are mentioned, run with no parameters — the script auto-infer
 .\release-notes\RunApiDiff.ps1 [mapped parameters]
 ```
 
-Set an initial wait of at least 300 seconds — the script takes several minutes. Use `read_powershell` to poll for completion. The script does not print a final "done" message; it exits after generating a README.md in the output folder. After completion, summarize the results: how many diff files were generated and where.
+Set an initial wait of at least 300 seconds — the script takes several minutes. After completion, summarize the results: how many diff files were generated and where.
@@ -0,0 +1,109 @@
+---
+name: editorial-scoring
+description: >
+  Apply the shared reader-centric rubric used to rank candidate features for
+  release notes, blog posts, and docs. Use this when you need scoring and cut
+  guidance independent of any one output format or task.
+---
+
+# Editorial Scoring
+
+Use this skill whenever you need to answer **"How important is this to a reader?"**
+
+This skill is intentionally **rubric-focused, not task-focused**. It does not generate files on its own. Instead, it provides the shared editorial calibration that other skills should reuse:
+
+- `generate-features` — assigns the first-pass scores
+- `review-release-notes` — audits and recalibrates those scores
+- `release-notes` — uses the resulting cut to decide what gets written up
+- future blog/docs workflows — can reuse the same rubric with different thresholds
+
+This is the **canonical home** for the shared scoring scale and 80/20 audience
+filter. Other release-notes docs should point back here instead of restating the
+rubric in full.
+
+## Core question
+
+Score from the perspective of a developer upgrading to the new release — not from the perspective of the engineer who implemented the feature.
+
+## Reader-centric scale
+
+| Score | Reader reaction                                                                  | What to do with it                                         |
+| ----- | -------------------------------------------------------------------------------- | ---------------------------------------------------------- |
+| `10`  | "This is the first feature I'll enable or test."                                 | Lead story                                                 |
+| `8+`  | "I'm going to use this when I upgrade."                                          | Strong release-note feature                                |
+| `6+`  | "I'm glad I know about this. It will likely come in handy."                      | Good grouped release-note material                         |
+| `4+`  | "I can see how someone could use this. I'll look up the docs if I ever need it." | Optional mention, grouping candidate, or short paragraph   |
+| `2+`  | "This one is a mystery to me."                                                   | Usually skip unless stronger explanation changes the score |
+| `0`   | "This is total gobbledygook — internal .NET engineering work."                   | Cut it from public-facing content                          |
+
+## 80/20 audience filter
+
+Default to features that make sense to roughly **80% of the audience**.
+
+Keep a more specialized feature for the other **20%** only when the remaining 80% can still understand why it matters and react with something like:
+
+> "Not for me, but I'm glad that's there for the people who need it."
+
+This is how foundational but initially niche work can still earn a good score.
+
+## Strong positive signals
+
+- New public APIs with obvious user value
+- CLI or workflow improvements developers will notice right away
+- Features that are easy to explain by anchoring them to a familiar tool or workflow
+- Clear performance, reliability, or usability wins
+- Features people have been asking for
+- Changes that require docs, migration notes, or examples to use well
+
+## Strong negative signals
+
+- Test-only work
+- Build or infrastructure churn
+- Refactoring with no user-visible effect
+- VMR sync noise, dependency updates, or repo automation
+- Highly specialized implementation details that read like internal jargon
+- Features that only matter after several stacked conditions are true (for example: uses single-file publish **and** cares deeply about startup **and** is willing to do extra training/tuning)
+- PRs with very thin descriptions where the only concrete signal is an internal runtime/tooling term like cDAC
+- Existing niche surfaces that are barely documented and are not part of the model's normal customer-facing understanding
+- Claimed "new API" features where the actual new public API cannot be identified
+- Bug fixes for features that were never really announced or are still obscure to most readers
+- Old, low-engagement bugs filed internally that sat for a long time without clear external demand
+
+## Common scoring mistakes
+
+- **Technical novelty bias** — "this is clever" is not the same as "users will care"
+- **API inventory mode** — listing everything from an API diff is not release-note curation
+- **Effort bias** — a difficult implementation may still score low if the reader value is small
+- **Insider-language inflation** — terms like JIT, LSRA, cDAC, intrinsics, or pipeline details can sound impressive but still be a `0-4` unless the value is explained clearly
+- **Audience multiplication blindness** — when a feature requires several independent interests or behaviors, multiply those filters together instead of scoring the broadest prerequisite alone
+- **Description vacuum optimism** — if the PR does not explain the user scenario, do not fill in the blanks with a generous story
+- **Documentation blindness** — if an "existing" feature is not findable in Learn docs and does not show up as a known customer scenario, that is a signal to score it around `1`, not to assume hidden importance
+- **Phantom API stories** — if you cannot identify the concrete new public API, do not give API credit based on a vague title alone
+- **Promotion by association** — a fix in a glamorous subsystem is still just a `1` if it only patches an unannounced or niche feature
+- **Demand inversion** — do not mistake an old internal bug with no reactions for evidence that many users need to hear about the fix
+- **Fragmentation bias** — do not dismiss each small related item in isolation when several `2-4` items together form one intelligible reader story, such as a cluster of "Unsafe evolution" changes
+
+## Working thresholds
+
+- **8+** — worthy of its own section
+- **6-7** — usually grouped with similar items, sometimes a short standalone section
+- **0-5** — bug-fix bucket, one-liner, or cut
+
+Several `3-5` items in the same area can still combine into **one** worthwhile writeup. Keep the individual scores honest, then let the writing stage roll them up into a single themed section.
+
+## Breaking changes are separate from score
+
+Use score for **reader interest/value**, and use `breaking_changes: true` for **reaction/migration significance**.
+
+That means:
+
+- a **low-score breaking change** often still deserves a short end-of-notes callout
+- a **high-score breaking change** may deserve both a full feature writeup and a migration note
+
+Do **not** inflate a narrow breaking change to `7+` just to keep it visible.
+
+## Canonical references
+
+- [`../release-notes/references/feature-scoring.md`](../release-notes/references/feature-scoring.md) — detailed scoring guidance
+- [`../release-notes/references/quality-bar.md`](../release-notes/references/quality-bar.md) — what good release notes look like
+- [`../release-notes/references/examples/README.md`](../release-notes/references/examples/README.md) — editorial examples and why they work
@@ -0,0 +1,129 @@
+---
+name: generate-changes
+description: >
+  Generate `changes.json` for a .NET release milestone by selecting the correct
+  VMR base/head refs and running `release-notes-gen generate changes`. Handles
+  preview-only multi-branch targeting (`main` vs release branches vs tags) and
+  emits the authoritative manifest of what shipped. DO NOT USE FOR: API
+  verification/diffs (use api-diff), feature scoring (use generate-features),
+  or writing markdown release notes (use release-notes).
+---
+
+# Generate `changes.json`
+
+Produce the authoritative `changes.json` input for a preview, RC, or GA release milestone.
+
+This is the **VMR-aware data acquisition stage** of the release notes pipeline:
+
+1. Determine which milestone(s) are active
+2. Resolve the correct `--base` and `--head` refs
+3. Run `release-notes-gen generate changes`
+4. Write `changes.json` into the correct `release-notes/` folder
+
+## When to use
+
+- A new preview, RC, or GA milestone needs fresh shipped-change data
+- The user wants to know whether **multiple preview milestones** are active at once
+- The release notes branch should be refreshed after the VMR moved forward
+- `changes.json` is missing, stale, or suspected to have the wrong ref selection
+
+## Preview-only branch targeting
+
+This is the subtle part, and it mostly matters for **previews**.
+
+Multiple preview milestones can be active simultaneously:
+
+```text
+Latest shipped in this repo: Preview 3
+VMR main:                    Preview 5
+VMR release branch exists:   Preview 4
+
+→ Generate one `changes.json` for Preview 4
+→ Generate one `changes.json` for Preview 5
+```
+
+For each target milestone `N`:
+
+| Milestone state                         | Base ref    | Head ref           |
+| --------------------------------------- | ----------- | ------------------ |
+| Tag exists for N                        | Tag for N-1 | Tag for N          |
+| Release branch exists for N, no tag yet | Tag for N-1 | Release branch tip |
+| Only on `main`                          | Tag for N-1 | `main`             |
+
+**Critical rule:** never use `main` for milestone `N` if `main` has already moved to `N+1`.
+
+## Inputs
+
+The user should provide as much of this as they know:
+
+- **Target release** — e.g. `.NET 11 Preview 4`, `.NET 10 RC 2`, `.NET 10 GA`
+- **VMR clone path** — defaults to a local clone of `dotnet/dotnet`
+- Optionally, the exact refs if they already know them
+
+If the user does **not** specify the milestone, infer it from:
+
+1. `release-notes/{version}/releases.json` in this repo
+2. `eng/Versions.props` on `main` in the VMR
+3. Matching preview tags and release branches in the VMR
+
+## Process
+
+### 1. Determine the floor from `releases.json`
+
+Find the latest shipped milestone in this repo. That tells you the lowest in-flight milestone that may need work.
+
+### 2. Inspect the VMR
+
+- Read `eng/Versions.props` on `main` to determine the current prerelease iteration
+- List matching VMR tags for finalized milestones
+- List matching VMR release branches for stabilizing milestones
+
+Use the [VMR structure reference](../release-notes/references/vmr-structure.md) for naming conventions and branch patterns.
+
+### 3. Resolve `--base` and `--head`
+
+For each active milestone:
+
+- `--base` is the previous shipped milestone tag
+- `--head` is the milestone tag, release branch tip, or `main`, depending on what exists
+
+### 4. Generate the file
+
+```bash
+release-notes-gen generate changes <vmr-clone-path> \
+  --base <previous-release-tag> \
+  --head <current-release-ref> \
+  --version "<release-version>" \
+  --date "<yyyy-mm-dd>" \
+  --labels \
+  --output release-notes/<major.minor>/<milestone-path>/changes.json
+```
+
+Examples:
+
+```bash
+# Preview milestone
+release-notes-gen generate changes ~/git/dotnet \
+  --base v11.0.0-preview.3.26210.100 \
+  --head origin/release/11.0.1xx-preview4 \
+  --version "11.0.0-preview.4" \
+  --labels \
+  --output release-notes/11.0/preview/preview4/changes.json
+
+# GA/patch milestone
+release-notes-gen generate changes ~/git/dotnet \
+  --base v10.0.7 \
+  --head v10.0.8 \
+  --version "10.0.8" \
+  --output release-notes/10.0/10.0.8/changes.json
+```
+
+## Output contract
+
+The output file must follow the shared schema documented in [changes-schema.md](../release-notes/references/changes-schema.md):
+
+- top-level `release_version`, `release_date`, `changes`, `commits`
+- stable `id` values in `repo@shortcommit` format
+- same authoritative source of truth used by later skills
+
+Once `changes.json` exists, the next step is usually `generate-features`.