fix(lint): reword the two Vale errors this branch put where nothing lints

`Kyberforge.SentenceOpenerThereIs` is `level: error`, but `.vale.ini` scopes the
style to `[**/SKILL.md]`, so no hook can see a violation in `references/`. This
branch's own remedy — move prose out of the body into `references/` — is what
carried these two there:

- `provider-adapter-author/references/provider-matrix.md:31`
- `agent-audit/references/finding-criteria.md:95`

Both confirmed by copying the file to `SKILL.md` and running the audit's own
`vale-wrap.sh`: 1 error each before, 0 after. Reworded to name the subject
directly, which is what the rule asks for.

Five further occurrences exist in `references/` files already on `main`. Those
are the pre-existing corpus, not this branch's regression, and stay with #117 —
which is the real fix: widen the Vale scope so the gate can see this class at
all, and make the wrapper distinguish "0 alerts in 1 file" from "0 files
matched".

Source and generated mirror move together so the commit passes
`check-plugin-content-sync` standalone.

Refs: #117

Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01Ktx4QJzTXZtw35m6T9WhcK

plugins/core/.apm/skills/provider-adapter-author/references/provider-matrix.md

@@ -28,5 +28,5 @@ This skill is reached two ways: invoked directly by a user, and composed into by
 once it has written or updated the repo's `AGENTS.md`. Behave identically either way — do not
 assume a caller skill exists. Detect the provider file yourself, confirm `AGENTS.md` yourself, and
 run the closeout validator yourself, rather than treating any step as already done by the caller or
-as something the caller will do afterwards. There is no handshake to rely on and no state passed
-in beyond the file paths.
+as something the caller will do afterwards. No handshake exists to rely on, and no state is
+passed in beyond the file paths.

plugins/core/skills/provider-adapter-author/references/provider-matrix.md

@@ -28,5 +28,5 @@ This skill is reached two ways: invoked directly by a user, and composed into by
 once it has written or updated the repo's `AGENTS.md`. Behave identically either way — do not
 assume a caller skill exists. Detect the provider file yourself, confirm `AGENTS.md` yourself, and
 run the closeout validator yourself, rather than treating any step as already done by the caller or
-as something the caller will do afterwards. There is no handshake to rely on and no state passed
-in beyond the file paths.
+as something the caller will do afterwards. No handshake exists to rely on, and no state is
+passed in beyond the file paths.

plugins/kyberforge/.apm/skills/agent-audit/references/finding-criteria.md

@@ -92,7 +92,7 @@ Flag as SUGGESTION if:
 
 **Never report an agent body as too long on a word count.** ADR-0020 gates a skill body at
 600/900 words and deliberately gates an agent body at nothing, because an agent body *becomes* the
-system prompt of a fresh context rather than competing with a live conversation. There is no number
+system prompt of a fresh context rather than competing with a live conversation. No number exists
 to cite. The one length signal that applies is the Copilot runtime's 30,000-character body limit,
 which `validate.sh` already reports as a SUGGESTION. Length is judged through the delegation FAIL
 above instead.

plugins/kyberforge/skills/agent-audit/references/finding-criteria.md

@@ -92,7 +92,7 @@ Flag as SUGGESTION if:
 
 **Never report an agent body as too long on a word count.** ADR-0020 gates a skill body at
 600/900 words and deliberately gates an agent body at nothing, because an agent body *becomes* the
-system prompt of a fresh context rather than competing with a live conversation. There is no number
+system prompt of a fresh context rather than competing with a live conversation. No number exists
 to cite. The one length signal that applies is the Copilot runtime's 30,000-character body limit,
 which `validate.sh` already reports as a SUGGESTION. Length is judged through the delegation FAIL
 above instead.