From 14160f9317588c7b4550329290691fb021da6bc9 Mon Sep 17 00:00:00 2001
From: Hermes <hermes@nousagent.com>
Date: Thu, 8 Oct 2026 14:55:54 -0700
Subject: [PATCH] [grade=B] fix(DC-163 rev2): strict merge semantics,
 contract-proving settings/test, valid enginecheck fixture
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Judge r1 (C) blockers B1-B3 + polish both applied — see body of rev1 commit for feature scope.

urn:ump:l2tg66wmchkmd6fde4k3ojwc56zpxqh657spe37bu6dpvduswalq
---
 .../__tests__/routes/ai-intent-engine.test.js | 38 ++++++++++++++++
 dashcaddy-api/routes/ai-intent.js             | 44 ++++++++++++++-----
 2 files changed, 71 insertions(+), 11 deletions(-)

diff --git a/dashcaddy-api/__tests__/routes/ai-intent-engine.test.js b/dashcaddy-api/__tests__/routes/ai-intent-engine.test.js
index c5a4568..32441a3 100644
--- a/dashcaddy-api/__tests__/routes/ai-intent-engine.test.js
+++ b/dashcaddy-api/__tests__/routes/ai-intent-engine.test.js
@@ -180,6 +180,27 @@ describe('DC-163 classify engine selection', () => {
     expect(out.appId).toBe('uptime-kuma'); // NOT in the old regex list — catalog-wide
   });
 
+  test('LLM reroute: pattern deploy result + LLM unknown → action null, stale appId dropped', async () => {
+    // Judge r1 blocker #1: when the LLM overrides the pattern's intent,
+    // NO intent-specific field from the pattern may ride along.
+    const cm = fakeCredStore(JSON.stringify({ baseUrl: 'http://llm.test/v1', apiKey: '***', model: 'test-model' }));
+    const out = await intentMod.classify({
+      message: 'Deploy Plex', // pattern: deploy + appId plex
+      history: [],
+      deps: { credentialManager: cm, env: {}, fetchImpl: llmReplyFetch('{"intent":"rm -rf","appId":"plex","reply":"hmm"}'), log: silentLog, timeoutMs: 5000 },
+    });
+    expect(out.engine).toBe('llm');
+    expect(out.intent).toBe('unknown'); // illegal value coerced
+    // action must be strictly null (unknown has no action) — never the
+    // pattern's deploy action riding along
+    expect(out.action).toBeNull();
+    // stale pattern appId must NOT be carried under the new intent
+    expect(out.appId === undefined || out.appId === null).toBe(true);
+    const enriched = intentMod.enrich(out);
+    expect(enriched.response.message).toBeTruthy(); // rebuilt for unknown, no deployPlan
+    expect(enriched.deployPlan).toBeUndefined();
+  });
+
   test('LLM disagreement: pattern said recommend, LLM says deploy → deploy response rebuilt', async () => {
     const cm = fakeCredStore(JSON.stringify({ baseUrl: 'http://llm.test/v1', apiKey: LLM_TEST_KEY }));
     const out = await intentMod.classify({
@@ -352,6 +373,23 @@ describe('DC-163 ai HTTP routes', () => {
     } finally { server.close(); }
   });
 
+  test('POST /ai/settings/test — sample MUST classify as backup; coerced unknown = failure', async () => {
+    // Judge r1 blocker #2: accepting a coerced 'unknown' would let a model
+    // that ignores our prompt "pass" the test. Strict expectation now.
+    const cm = fakeCredStore(JSON.stringify({ baseUrl: 'http://llm.test/v1', apiKey: LLM_TEST_KEY, model: 'test-model' }));
+    const fetchImpl = llmReplyFetch('{"intent":"chitchat","appId":null,"reply":"lol sure"}'); // illegal -> coerces to unknown
+    const app = buildRouter({ credentialManager: cm, fetchImpl });
+    const { server, base } = await start(app);
+    try {
+      const res = await fetch(`${base}/ai/settings/test`, { method: 'POST', headers: { 'Content-Type': 'application/json' }, body: '{}' });
+      const body = await res.json();
+      expect(res.status).toBe(200);
+      expect(body.ok).toBe(false);
+      expect(body.engine).toBe('llm-error');
+      expect(body.reason).toMatch(/expected 'backup', got 'unknown'/);
+    } finally { server.close(); }
+  });
+
   test('GET /ai/capabilities — engine reflects live configuration', async () => {
     const app = buildRouter();
     const { server, base } = await start(app);
diff --git a/dashcaddy-api/routes/ai-intent.js b/dashcaddy-api/routes/ai-intent.js
index bc9a7d8..3ce0c79 100644
--- a/dashcaddy-api/routes/ai-intent.js
+++ b/dashcaddy-api/routes/ai-intent.js
@@ -167,19 +167,34 @@ function validateParsed(parsed) {
 }
 
 // ─── Merge validated LLM fields onto the pattern result ────────────────────
-// Pattern stays the structural base (known-good shapes); LLM contributes the
-// semantic understanding. appId is never LOST silently when both engines
-// disagree — the LLM candidate wins if it is a valid catalog id, because the
-// LLM saw the full 78-app catalog while regex knows only its hardcoded list.
+// Same-intent: pattern stays the structural base (known-good shapes) and
+// the LLM adds semantic corrections (a valid catalog appId wins even when
+// regex missed it). Different-intent: the pattern's intent-SPECIFIC fields
+// (appId/serviceId/categories/response/action) belong to the OLD intent and
+// are dropped — merged fields come exclusively from the validated LLM
+// output, with `action` strictly ACTIONS_BY_INTENT[llm.intent] (null for
+// 'unknown'). Stale pattern data can never ride along under a new intent.
 function mergeIntent(patternResult, llm) {
-  const merged = { ...patternResult };
   const intentChanged = llm.intent !== patternResult.intent;
-  merged.intent = llm.intent;
-  merged.action = ACTIONS_BY_INTENT[llm.intent] || patternResult.action || null;
-  // If the LLM routed to a DIFFERENT intent than the pattern did, the
-  // pattern's canned response object belongs to the old intent — drop it
-  // and let enrich() rebuild the shape for the new one (with llmReply text).
-  if (intentChanged) merged.response = null;
+  if (intentChanged) {
+    // Fields only ride along when they are meaningful for the NEW intent
+    // (a 'plex' appId under intent 'unknown' is meaningless — drop it).
+    return {
+      intent: llm.intent,
+      confidence: 0.85,
+      message: patternResult.message,
+      action: ACTIONS_BY_INTENT[llm.intent] ?? null,
+      response: null,
+      appId: (llm.intent === 'deploy' && llm.appId) ? llm.appId : undefined,
+      serviceId: (llm.intent === 'diagnose' && llm.serviceId) ? llm.serviceId : undefined,
+      categories: llm.intent === 'recommend'
+        ? ((llm.categories && llm.categories.length) ? llm.categories : ['media-streaming'])
+        : undefined,
+      llmReply: llm.reply || undefined,
+    };
+  }
+  const merged = { ...patternResult };
+  merged.action = ACTIONS_BY_INTENT[llm.intent] ?? null;
   if (llm.appId) merged.appId = llm.appId;
   if (llm.serviceId && !merged.serviceId) merged.serviceId = llm.serviceId;
   if (llm.intent === 'recommend') {
@@ -645,6 +660,13 @@ module.exports = function ({ asyncHandler, siteConfig = null, log = console, fet
       const parsed = JSON.parse(repairParsed(content));
       const validated = validateParsed(parsed);
       if (!validated || !validated.intent) throw new Error('invalid classifier output');
+      // Contract proof: the fixed sample MUST classify as 'backup'. An
+      // 'unknown' here means the endpoint answers but the model/prompt is
+      // not actually doing DashCaddy classification — report it as a
+      // failure, not a pass.
+      if (validated.intent !== 'backup') {
+        return ok(res, { ok: false, engine: 'llm-error', reason: `sample misclassified: expected 'backup', got '${validated.intent}'` });
+      }
       return ok(res, { ok: true, engine: 'llm', model: cfg.model, sample: { intent: validated.intent, reply: validated.reply } });
     } catch (e) {
       const reason = String((e && (e.name === 'AbortError' ? 'timeout' : e.message)) || 'error').slice(0, 200);