feat: read CV data from plugin file instead of hardcoded defaults

_data/cv.js now reads from content/.indiekit/cv.json written by the
CV plugin, matching the homepageConfig.js pattern. Falls back to
empty defaults when no plugin is installed.

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>

_data/cv.js

@@ -1,16 +1,28 @@
 /**
- * CV Data — Empty defaults.
+ * CV Data — reads from indiekit-endpoint-cv plugin data file.
  *
- * When the indiekit-endpoint-cv plugin is installed, it serves CV data
- * via its API endpoint and the homepage plugin renders it.
+ * The CV plugin writes content/.indiekit/cv.json on every save
+ * and on startup. Eleventy reads that file here.
  *
- * Without the plugin, users can edit this file directly:
- *   - Add entries to the `experience` array
- *   - Add entries to the `projects` array
- *   - Modify the `skills` object
+ * Falls back to empty defaults if no plugin is installed.
  */
 
-export default {
+import { readFileSync } from "node:fs";
+import { resolve, dirname } from "node:path";
+import { fileURLToPath } from "node:url";
+
+const __dirname = dirname(fileURLToPath(import.meta.url));
+
+export default function () {
+  try {
+    const cvPath = resolve(__dirname, "..", "content", ".indiekit", "cv.json");
+    const raw = readFileSync(cvPath, "utf8");
+    const data = JSON.parse(raw);
+    console.log("[cv] Loaded CV data from plugin");
+    return data;
+  } catch {
+    // No CV plugin data file — return empty defaults
+    return {
       lastUpdated: null,
       experience: [],
       projects: [],
@@ -19,3 +31,5 @@ export default {
       education: [],
       interests: [],
     };
+  }
+}