feat(newos4): world-model-ask answers with the LLM, grounded in the model (v4.6.50)

`world-model-ask` was a 10-keyword lookup: any real question returned
"Unknown query. Try: modules, skills, ...". The documented example
("what changed since yesterday") compared nothing — it restated the current
snapshot.

- _modelDigest(): whitelisted snapshot (measured 942-965 chars on fixtures,
  ~3.3 KB live) + recent error sample + changesSinceLastSnapshot
- _buildAskPrompt(): "answer only from this JSON, never invent numbers, name the
  command that would produce missing data"
- ask(): connects Ollama on demand, answers via chat(), falls back to the facts
  path with a visible warning; --facts keeps the instant lookup
- _diffSincePrev(): keeps one previous snapshot in /system/world-model.prev.json
  so "what changed" is a real diff (files 136 -> 140)
- world_model_query tool returns the snapshot (digest or detail:full) instead of
  ten canned strings; no nested LLM call inside a tool call
- help text now says what it does; node test modules/test/test-world-model-ask.js
  (11 assertions, runs the real class)

Verified live v4.6.50: facts "Loaded 46 modules" | LLM "The snapshot does not
contain information regarding the cause of memory growth ... use
world-model-refresh" | diff "files 136 -> 140" | tool snapshot 3,334 bytes.

1 files changed, 47 insertions, 0 deletions

diff --git a/modules/test/test-world-model-ask.js b/modules/test/test-world-model-ask.js
new file mode 100644
index 0000000..4a0c21a
--- /dev/null
+++ b/modules/test/test-world-model-ask.js
@@ -0,0 +1,47 @@
+// node modules/test/test-world-model-ask.js
+// Executes the real class from module-world-model.js (no imports, no DOM at class level).
+const fs = require('fs');
+const path = require('path');
+const src = fs.readFileSync(path.join(__dirname, '../core/module-world-model.js'), 'utf8');
+const WorldModelModule = new Function(src + '\nreturn WorldModelModule;')();
+
+const wm = new WorldModelModule();
+const failures = [];
+const check = (name, cond, detail) => {
+    if (!cond) failures.push(name + (detail ? ' — ' + detail : ''));
+    console.log((cond ? 'PASS' : 'FAIL'), name, detail || '');
+};
+
+// no model yet
+check('empty model digest is empty', wm._modelDigest() === '');
+
+wm._model = {
+    version: 'v4.6.47', synthesized: '2026-10-21T20:00:00.000Z',
+    system: { health: 'ok', memory: '3 MB' },
+    modules: { loaded: 30, total: 30, modifiedCount: 0, modified: [] },
+    files: { total: 42, directories: 8, size: '12 KB' },
+    skills: { count: 23, skills: Array.from({ length: 60 }, (_, i) => 'skill-' + i) },
+    memory: { nodes: 120, edges: 200, active: true },
+    agents: { agents: 4, active: 1 },
+    telemetry: { toolLatencyDataPoints: 9, recentErrors: 0 },
+    projects: { totalTasks: 3, activeProjects: [] },
+    recent: ['Skills directory: 23 files']
+};
+
+const digest = wm._modelDigest();
+check('digest is valid JSON', (() => { try { JSON.parse(digest); return true; } catch (e) { return false; } })());
+check('digest is capped at 4000', digest.length <= 4000, digest.length + ' chars');
+check('digest keeps counts', /"count":23/.test(digest) && /"loaded":30/.test(digest));
+check('digest trims the skill list to 40', (JSON.parse(digest).skills.skills || []).length === 40,
+      String((JSON.parse(digest).skills.skills || []).length));
+
+const prompt = wm._buildAskPrompt('  why did the build break?  ');
+check('prompt carries the question', /QUESTION: why did the build break\?$/.test(prompt), prompt.slice(-60));
+check('prompt carries the snapshot', prompt.indexOf('SNAPSHOT: {') > 0);
+check('prompt forbids invention', /Never invent numbers/.test(prompt));
+
+check('facts path still answers a keyword', /30 modules/.test(wm._answerQuery('how many modules')));
+check('facts path still refuses nonsense', /Unknown query/.test(wm._answerQuery('why did it break')));
+
+console.log(failures.length ? '\n' + failures.length + ' FAILED' : '\nall PASS');
+process.exit(failures.length ? 1 : 0);