crates/queryd/src/service.rs

@@ -12,8 +12,6 @@ use serde::{Deserialize, Serialize};
 use crate::context::QueryEngine;
 use crate::delta;
 use crate::paged::ResultStore;
-use tracing;
-use std::collections::HashSet;
 
 #[derive(Clone)]
 pub struct QueryState {

mcp-server/observer.ts

@@ -159,63 +159,50 @@ const LLM_TEAM_ESCALATIONS = "/home/profit/lakehouse/data/_kb/observer_escalatio
 const ESCALATION_THRESHOLD = 3; // N+ failures on same sig_hash triggers
 
 async function escalateFailureClusterToLLMTeam(sigHash: string, cluster: ObservedOp[]) {
-  // Package the failure cluster as a single context blob. Originally
+  // Package the failure cluster as a single context blob for code_review mode.
-  // I routed this to LLM Team's `code_review` mode at /api/run, but
-  // that mode isn't registered in llm_team_ui.py — it returned
-  // "Unknown mode" on every call. Revised 2026-04-24: route directly
-  // to the gateway's /v1/chat with provider=ollama_cloud + qwen3-coder:480b
-  // (the coding specialist that's rung 2 of the scrum ladder, proven
-  // to produce substantive structured reviews). Fire-and-forget so
-  // downstream failures don't block observer's normal loop.
   const context = cluster.slice(-8).map((o, i) =>
     `[${i + 1}] endpoint=${o.endpoint} input=${o.input_summary} error=${o.error ?? "?"}`
   ).join("\n");
-  const prompt = `sig_hash=${sigHash} · ${cluster.length} failures on the same signature:\n\n${context}\n\nReview this failure cluster. Identify:\n1. Likely root cause (single sentence).\n2. Files most likely responsible (path hints).\n3. Concrete fix direction (under 3 sentences).\n4. Confidence: NN%\n\nBe specific, not generic.`;
 
   try {
-    const resp = await fetch(`${LAKEHOUSE}/v1/chat`, {
+    const resp = await fetch(`${LLM_TEAM}/api/run?mode=code_review`, {
       method: "POST",
       headers: { "Content-Type": "application/json" },
       body: JSON.stringify({
-        provider: "ollama_cloud",
+        input: `sig_hash=${sigHash} · ${cluster.length} failures on same signature:\n\n${context}\n\nReview this failure pattern. What is the root cause? What code change would prevent it? Respond with structured facts + specific file hints.`,
-        model: "qwen3-coder:480b",
-        messages: [{ role: "user", content: prompt }],
-        max_tokens: 800,
-        temperature: 0.2,
       }),
       signal: AbortSignal.timeout(60000),
     });
     if (!resp.ok) {
-      console.error(`[observer] escalation /v1/chat ${resp.status}: ${(await resp.text()).slice(0, 200)}`);
+      console.error(`[observer] LLM Team code_review ${resp.status}: ${(await resp.text()).slice(0, 200)}`);
       return;
     }
     const j: any = await resp.json();
-    const analysis = j?.choices?.[0]?.message?.content ?? "";
 
-    // Audit row stays schema-compatible with the prior implementation —
+    // Write an audit row. Fields are deliberately permissive — LLM
-    // downstream consumers see structured fields regardless of the
+    // Team's response shape can evolve without breaking this write.
-    // review-source change. Facts/entities stay empty (this call is
-    // direct-model, not extract-mode); the raw analysis carries the
-    // signal.
     const row = {
       ts: new Date().toISOString(),
       source: "observer_escalation",
-      mode: "direct_chat_qwen3_coder_480b",
+      mode: "code_review",
       sig_hash: sigHash,
       cluster_size: cluster.length,
       cluster_staffer: cluster[0]?.staffer_id,
       cluster_endpoint: cluster[0]?.endpoint,
-      prompt_tokens: j?.usage?.prompt_tokens ?? 0,
+      llm_team_run_id: j.run_id ?? j.llm_team_run_id ?? null,
-      completion_tokens: j?.usage?.completion_tokens ?? 0,
+      facts: j.facts ?? [],
-      analysis: analysis.slice(0, 4000),
+      entities: j.entities ?? [],
+      relationships: j.relationships ?? [],
+      raw_response: typeof j.response === "string" ? j.response.slice(0, 2000) : null,
+      recommended_files: j.file_hints ?? j.files ?? [],
     };
     const { appendFile } = await import("node:fs/promises");
     await appendFile(LLM_TEAM_ESCALATIONS, JSON.stringify(row) + "\n");
     console.error(
-      `[observer] escalated sig_hash=${sigHash.slice(0, 8)} · cluster=${cluster.length} · ${analysis.length} chars`
+      `[observer] escalated sig_hash=${sigHash.slice(0, 8)} · cluster=${cluster.length} · facts=${row.facts.length} entities=${row.entities.length}`
     );
   } catch (e) {
-    console.error(`[observer] escalation failed: ${(e as Error).message}`);
+    console.error(`[observer] LLM Team escalation failed: ${(e as Error).message}`);
   }
 }
 

tests/real-world/scrum_applier.ts

@@ -35,14 +35,7 @@ const AUDIT_LOG = `${REPO}/data/_kb/auto_apply.jsonl`;
 const MIN_CONF = Number(process.env.LH_APPLIER_MIN_CONF ?? 90);
 const MAX_FILES = Number(process.env.LH_APPLIER_MAX_FILES ?? 5);
 const COMMIT = process.env.LH_APPLIER_COMMIT === "1";
-// Default patch-emitter model — qwen3-coder:480b is the coding specialist
+const MODEL = process.env.LH_APPLIER_MODEL ?? "kimi-k2:1t";
-// in the scrum ladder (rung 2). Swapped in from kimi-k2:1t after 2026-04-24
-// data showed kimi-k2:1t produces architectural patches that cascade across
-// the file (rename a field → 20 broken call sites). qwen3-coder is tuned
-// for targeted code changes and tends to stay within the mechanical-patch
-// constraint the prompt asks for. LLM Team's /api/run?mode=patch would be
-// the ideal choice but that mode isn't registered in llm_team_ui.py yet.
-const MODEL = process.env.LH_APPLIER_MODEL ?? "qwen3-coder:480b";
 const BRANCH = process.env.LH_APPLIER_BRANCH ?? `scrum/auto-apply-${Date.now().toString(36)}`;
 
 // Deny-list — anything whose path starts with one of these is skipped
@@ -168,25 +161,14 @@ ${review}
 
 The review is high-confidence and the file is eligible for auto-apply. Produce CONCRETE PATCHES as JSON so they can be applied via string replacement.
 
-HARD CONSTRAINTS (violations → patch rejected):
+RULES:
   1. Output ONE JSON object with a "patches" array. NO prose, no markdown fences.
   2. Each patch is {"old_string": "...", "new_string": "...", "rationale": "short", "confidence": 0-100}.
-  3. "old_string" MUST appear EXACTLY ONCE in the file (verbatim, including whitespace + trailing newlines).
+  3. "old_string" MUST appear EXACTLY ONCE in the file (verbatim, including whitespace). If no unique anchor exists, SKIP that suggestion.
-  4. Max diff size: 6 lines changed per patch (NOT 20). If the change needs >6 lines, emit {"patches": []} — too risky for auto-apply.
+  4. Mechanical changes only: wire a function call, add a field, remove #[allow(dead_code)], add a missing use import, rename one call-site. NO architectural rewrites. NO new modules.
-  5. Mechanical-only (the list — nothing else):
+  5. Each "new_string" MUST compile in isolation with the same surrounding code. Don't introduce new dependencies.
-       (a) remove a #[allow(dead_code)] marker on a function now wired elsewhere
+  6. If you cannot produce at least one high-confidence mechanical patch, output {"patches": []}.
-       (b) add a missing 'use' import statement
+  7. Max 3 patches per file.
-       (c) add a single field to a struct (no renames)
-       (d) flip a single boolean/match-arm that doesn't cascade
-       (e) add a fire-and-forget log or tracing call
-  6. FORBIDDEN (automatic reject):
-       • struct field RENAMES (break all call sites)
-       • function signature changes
-       • new modules, new traits, new dependencies
-       • any change that requires editing another file to keep it compiling
-  7. Each "new_string" MUST compile in isolation with the same surrounding code.
-  8. Max 2 patches per file — quality over quantity.
-  9. If you cannot produce at least one high-confidence mechanical patch under these constraints, output {"patches": []}. Don't guess.
 
 ─── SOURCE (${source.length} bytes) ───
 ${source.slice(0, 14000)}
@@ -241,12 +223,10 @@ async function applyPatches(file: string, patches: Patch[]): Promise<{ applied:
     const occurrences = source.split(p.old_string).length - 1;
     if (occurrences === 0) { rejected.push({patch: p, reason: "old_string not found"}); continue; }
     if (occurrences > 1) { rejected.push({patch: p, reason: `old_string appears ${occurrences}× (not unique)`}); continue; }
-    // Size gate — no patch touches > 6 lines (diff discipline; matches prompt).
+    // Size gate — no patch touches > 20 lines (diff discipline).
-    // Raised from 20 to 6 after 2026-04-24 data showed most 10-20 line patches
-    // cascaded and broke the build. Mechanical changes genuinely fit in 6 lines.
     const oldLines = p.old_string.split("\n").length;
     const newLines = p.new_string.split("\n").length;
-    if (Math.max(oldLines, newLines) > 6) { rejected.push({patch: p, reason: `patch too large (${Math.max(oldLines,newLines)} lines, max 6)`}); continue; }
+    if (Math.max(oldLines, newLines) > 20) { rejected.push({patch: p, reason: `patch too large (${Math.max(oldLines,newLines)} lines)`}); continue; }
     source = source.replace(p.old_string, p.new_string);
     applied++;
   }