refactor(skills): retrofit the corpus to the ADR-0020 context contract (#129)

Retrofits all 39 skills to ADR-0020's description/body context contract, then fixes what six rounds of independent review found in that retrofit — including four ways the hot gate itself failed open.

Closes #99, #107, #108, #110, #111, #114, #115, #120.

## The retrofit (waves 1-5)

| | Start | Now |
|---|---|---|
| Description FAILs (>400 chars) | 26 | **0** |
| Body FAILs (>900 words, body-only) | 9 | **0** |
| Dangling routing targets | 2 | **0** |
| `Kyberforge.CompositionNote` | 10 | **0** |
| Preload tax | 21,005 chars | **~10,500** |

Under the 12,000-char success criterion. Per-wave detail is on #99.

## The review fixes

**The gate failed open four ways, three of them found after the retrofit shipped.** An unrecognised follower token made a dangling target vanish. A skill directory with no `SKILL.md` resolved as a valid target, so a commit could be green locally and red in a fresh clone — three existing fixtures were relying on that, one of which made the install-leak A/B pass vacuously. Then the free-standing `/name` sweep turned out to be gated on the sentence carrying a boundary marker, so route notation in any other sentence was invisible — not an ERROR, not a SUGGESTION, not an INFO — which left the documented "`/name` always blocks" promise false from a second direction. All four fixed and pinned.

**Two checks were silently not running.** `validate-provenance.sh` checks 7-8 were dead across nine skills. Waking them exposed a deeper problem: they assume `Research doc:` names a source index, but 30 of 121 entries point at topic content documents, so every new check-7 INFO was a false positive and check 8 was saved from a false-FAIL flood only by an *unannounced* skip. Checks 7/8 are now scoped to source indexes and every skip announces itself (#121).

**The retrofit's own anti-goal, four times.** ADR-0020 warns that a blunt gate gets satisfied by deleting content rather than relocating it. `diagnose` and `skill-audit` relocated prose and then read it unconditionally; `prototype` and `vale-config` deleted rules outright that survived nowhere. All four addressed.

## Verification

- `bash tests/run-tests.sh --strict` — 24 suites, 0 skipped, 0 failed
- `bash tests/run-bats.sh` — 325 tests, 0 failures
- `pre-commit run --all-files` — 17/17
- `pre-commit run --hook-stage pre-push --all-files` — 16/16, with `apm marketplace check` and `apm pack --check-clean` run against the remote, not skipped
- `scripts/skill-size-check.sh` over all 39 skills — rc 0, 0 ERROR/FAIL, SUGGESTION-only
- Preload tax measured at **10,498 chars**, max description 390 — both inside budget
- Every new test proven non-vacuous by a deliberate mutation of the behaviour it covers

**Per-commit sync, stated accurately:** the ten commits from the latest review round each pass `check-plugin-content-sync` in isolation, verified by checking each out in a detached worktree with a clean between. The earlier gitea window (`dfacf05..bedbd1d`, nine commits) does **not** — its mirror was regenerated in one batch at `bbc7300`. An earlier revision of this description claimed the property held for every commit; it does not, and a bisect through that window lands on a red commit. **Squash-merge** to collapse it, or accept that this range is not bisectable.

## Version bump

Six plugins and the catalog take a **patch**, not a minor. The branch is **89 commits — 40 `fix` / 30 `refactor` / 12 `docs` / 5 `chore` / 2 `test` — zero `feat`, zero `!`, zero `BREAKING CHANGE`** — and adds no skill, agent, command or hook. (Two earlier revisions of this section cited a stale histogram, most recently 78 commits; the figures above are measured at HEAD.) Both rules this repo ships (`forge/references/version-bump.md`, landing in this PR, and `git-commits/references/conventional-commits-spec.md`) make that a patch, and the catalog set is unchanged at 7 entries.

Not settled by that: four published files were removed from the installed tree, three moved, and `caveman` gained `disable-model-invocation`, retiring its old triggers. Under a strict reading those are major-class and currently ship under `refactor:` with no marker. Whether the deployed skill surface is a public contract is written down nowhere — worth deciding, but it outlives this PR.

## Deliberately not in scope

#112 (cherry-pick ownership, now resolved in favour of `git-commits`), #113 (`rtk git` normalisation), #116 (research fan-out), #101 (audit-skill merge), #122 (non-spec skill-root files), #123 (no PRD producer) stay open. #117 is the one worth reading: the contract's remedy is to move prose into `references/`, which is exactly where neither the size gate nor Vale looks — and the blind spot is wider than #117 currently records, since there is no root `.vale.ini` at all, so every ADR, `CONTEXT.md` and `README.md` is unlinted too.

That blind spot let this branch carry two `level: error` `Kyberforge.SentenceOpenerThereIs` violations into `references/` files it created — `provider-adapter-author/references/provider-matrix.md:31` and `agent-audit/references/finding-criteria.md:95`. Both are reworded in `afadaae`, confirmed by routing each file through the audit's own `vale-wrap.sh` (1 error each before, 0 after). Five further occurrences sit in `references/` files already on `main`; those are the pre-existing corpus and stay with #117, which is the real fix.

Also unfixed and not this PR's: `apm install` appends a duplicate `SessionStart` entry to `.claude/settings.json`, so a fresh clone cannot get pre-push green without an edit AGENTS.md warns against. Reproduces identically on `main`.

Co-authored-by: Defame1297 <gitea@rkdr.net>
Reviewed-on: https://git.dev.rkdr.net/Defame1297/holocron/pulls/129
Co-authored-by: Claude Code AI - Gitea MCP <claude@noreply.git.dev.rkdr.net>
Co-committed-by: Claude Code AI - Gitea MCP <claude@noreply.git.dev.rkdr.net>

plugins/lint/.apm/skills/vale-config/references/configuration-reference.md

@@ -2,6 +2,7 @@
 topic: configuration-reference
 source_keys:
   - context7-websites-vale-sh
+  - house-vale-3-15-2-repro
 ---
 
 ## Core Settings
@@ -24,6 +25,10 @@ Map an unrecognized extension onto a supported one so Vale lints it with the rig
 mdx = md
 ```
 
+`mdx` is the case that matters, because Vale 3.15.2 has no built-in MDX support. The mapping above is not cosmetic: it is what lets `.mdx` files lint with nothing else installed. Leave it out and Vale takes the native MDX path, which shells out to an external `mdx2vast` binary — absent from `PATH`, the run dies with `E100 [lintMDX] Runtime error / mdx2vast not found`, exit 2, and every other file in the same invocation goes unlinted too. Take the mapping: it is this skill's recommended default, because it needs nothing installed, and it is the branch `vale-run` assumes when it documents inline suppressions. Install `mdx2vast` (`npm install -g mdx2vast`) only when something else in the toolchain already needs the native MDX parser.
+
+The choice also decides the inline-suppression syntax, and it is inverted between the two: mapped to `md`, `.mdx` takes Markdown's `<!-- vale off -->`; native, it takes `{/* vale off */}`. `vale-run`'s `references/troubleshooting.md` carries the verified matrix.
+
 ## Vocabularies
 
 Reference a named vocabulary (a folder of accept/reject word lists under `StylesPath`) via `Vocab`, then apply styles per glob:
@@ -68,11 +73,41 @@ Individual rule YAML files (under a style's directory) support these header fiel
 
 The underlying functions a rule's `extends` field can reference: `existence`, `substitution`, `occurrence`, `repetition`, `consistency`, `conditional`, `capitalization`, `metric`, `spelling`, `sequence`, `script`.
 
-## Built-in Style
+## Style Resolution
 
-Vale ships with a default `Vale` style containing four rules, usable without `vale sync`:
+Only *package* styles need fetching. A style whose YAML rule files are already committed under `StylesPath` lints immediately, with no `Packages` entry and no `vale sync`; the same is true of the built-in `Vale` style, which ships with the binary and contains four rules:
 
 - `Vale.Spelling` — spell-checks against Hunspell-compatible dictionaries in `<StylesPath>/config/dictionaries`.
 - `Vale.Terms` — enforces the project's accepted vocabulary terms.
 - `Vale.Avoid` — enforces the project's rejected vocabulary terms.
 - `Vale.Repetition` — flags repeated words (e.g. "the the").
+
+`Packages` (top-level, what `vale sync` downloads) and `BasedOnStyles` (per-glob, what activates) are separate keys: a style lints a file only once it is in both. Every row below reproduced against Vale 3.15.2 (slug `house-vale-3-15-2-repro`):
+
+| Configuration | Result |
+|---|---|
+| `BasedOnStyles` names a style with no directory under `StylesPath`, not built-in | `E100 [loadStyles] Runtime error` — `style 'X' does not exist on StylesPath`, exit 2 |
+| `StylesPath` directory itself absent, even with only `Vale` active | `E201 Invalid value` — `The path '...' does not exist`, exit 2 |
+| `vale sync` with a name in `BasedOnStyles` but not `Packages` | `SUCCESS Synced 0 package(s)`, exit 0, nothing downloaded — the next lint repeats the `E100` |
+| `vale sync` with the name added to `Packages` | package lands under `StylesPath`, exit 0; lint then loads it |
+| Style in `Packages` and synced, but in no glob's `BasedOnStyles` | 0 findings, exit 0 — downloads, never lints, indistinguishable from a clean run |
+| `BasedOnStyles` names an *empty* directory under `StylesPath` | 0 findings, exit 0 — loads and lints nothing; `vale sync` never produces this state |
+| Built-in `Vale`, or a style's YAML committed under `StylesPath` | lints immediately, no `Packages` entry, no sync |
+| Core option (`StylesPath`, `MinAlertLevel`, `Vocab`, `IgnoredScopes`, `SkippedScopes`) below a `[glob]` header | `E201 Invalid value` — `'X' is a core option; it should be defined above any syntax-specific options ([...])`, exit 2 |
+| `Packages` below a `[glob]` header | no error, exit unaffected — parsed as a per-glob rule toggle (`SChecks: {"*.md": {"Packages": false}}` in `ls-config`), so `vale sync` reports `Synced 0 package(s)` and downloads nothing |
+
+## Frontmatter Scopes
+
+House-verified behaviour, not documented on vale.sh — reproduced locally against Vale 3.15.2 (slug `house-vale-3-15-2-repro`).
+
+A rule scoped to `text.frontmatter.<key>` (e.g. `text.frontmatter.description`) matches reliably when that field's value is a single physical line, and breaks on most — not all — multi-line forms. Multi-line forms spanning 2+ lines:
+
+| Frontmatter value form | Result |
+|---|---|
+| Single physical line (control) | Lints, exits 1 |
+| `\|` literal block scalar | Lints, exits 1 |
+| `>` folded block scalar | 0 findings, exits 0 |
+| Plain (unquoted) continuation lines | 0 findings, exits 0 |
+| Single- or double-quoted multi-line scalar | 0 findings, exits 0 |
+
+The silent cases produce no error of any kind, so a passing run is indistinguishable from a clean one. Do not assume a literal block scalar and a folded one behave alike — reproduce both against your own config before trusting a frontmatter-scoped rule in production. If the field is commonly authored in one of the broken forms, flatten it to one physical line ahead of the `vale` call rather than relying on the scope alone.