chore: consolidate the skill packages under .agents/

The repository had two skill directories. `skills/` held the four written
for this project; `.agents/skills/` held the ones the agent context refers
to. Nothing said which an agent should read, and `.agents/ORCHESTRATOR.md`
only ever pointed at the second.

Move the first four into `.agents/skills/` so there is one location, and
add the vendored packages this chapter work used: `animation-vocabulary`
and `improve-animations` (emilkowalski/skills), `teach` (mattpocock/skills),
plus a local `translation` skill and `audit-translations.mjs` for the EN/PT
pairs.

`skills-lock.json` pins the vendored three by source and content hash, so a
later re-vendor is a diff rather than a guess. `.claude/skills/` is
symlinks into `.agents/skills/`, which is what makes them loadable here
without a second copy on disk.

Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>

.agents/skills/translation/evals/evals.json

@@ -0,0 +1,10 @@
+[
+  {
+    "prompt": "The pages /summary/ and /agents/ in this Astro bilingual site ship identical strings in their rendered Portuguese and English. Where do you start, and what is the smallest change that would expose the regression?",
+    "expected_behavior": "Run node .agents/scripts/audit-translations.mjs to find every identical localized field; pick the smallest chapter; load the translation skill (glossary.md and tone.md) before proposing candidates; show a diff for human review before writing."
+  },
+  {
+    "prompt": "A reviewer rejected your first chapter's Portuguese with 'this doesn't sound like the rest of the site'. What do you do next?",
+    "expected_behavior": "Compare rejected samples against src/content/commonSkills/*.json and src/content/rules/copy.json pt block; check the rejected entry against references/glossary.md term-by-term; check the rejected entry against references/tone.md (register, verb mood, capitalization, punctuation); do not commit and do not move to the next chapter until tone converges."
+  }
+]