fix(search): match both slash-split and literal slash forms in FTS queries (#2380 review)

Review finding on the normalize-only approach: Postgres' text-search
parser emits the joined file-alias lexeme on the INDEX side too
(to_tsvector('english','acme/widget') -> 'acme/widget'), so replacing
'/' with whitespace in the query made documents containing the literal
slash form (file paths, URLs, pasted titles) unreachable — the split-word
AND pass can't match the joined lexeme and the OR fallback can't either.
Pre-fix, those exact-form queries DID match.

Replace the TS-side normalizeKeywordQuery with buildWebsearchQueryExpr
in sql-ranking.ts (shared by both engines, keeping them in lockstep): a
slash-bearing query now binds the raw text once and matches
(websearch_to_tsquery(translate($1,'/',' ')) || websearch_to_tsquery($1))
— split-word prose AND literal slash forms alike. Slash-free queries keep
the byte-identical single-parse SQL and identical ts_rank.

New regression test pins the literal-slash arm (verified failing under
the normalize-only expression); the two AND-vs-OR slash tests still pass.

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

src/commands/doctor.ts

@@ -820,6 +820,10 @@ export async function doctorReportRemote(engine: BrainEngine): Promise<DoctorRep
   // v0.42.x (#1794, 4A): pool-budget nudge when GBRAIN_MAX_CONNECTIONS is set.
   checks.push(await checkPoolBudget(engine));
 
+  // #2552: warn when an explicit embed-concurrency override fans out against
+  // a local single-slot embedding endpoint (silent backfill starvation).
+  checks.push(await checkEmbedConcurrency());
+
   // v0.42.7 (#1696): link-extraction lag. Strictly SQL (single indexed COUNT),
   // safe on the thin-client/remote path — remote operators on checkout-less
   // Postgres brains are exactly who can't otherwise see the extraction backlog.
@@ -3815,6 +3819,61 @@ export function computePoolBudgetCheck(
   };
 }
 
+/**
+ * #2552: warn when an explicit GBRAIN_EMBED_CONCURRENCY override fans out
+ * against a local single-slot embedding endpoint (Ollama / llama-server /
+ * localhost base URL). Requests serialize on the one loaded model, so N
+ * parallel pages multiply latency xN and can exceed the fetch timeout with
+ * no surfaced error — the backfill silently starves. (When the env var is
+ * unset, embed auto-caps at LOCAL_EMBED_CONCURRENCY_CAP and this check
+ * reports ok.) Pure; exported for tests.
+ */
+export function computeEmbedConcurrencyCheck(
+  isLocalEndpoint: boolean,
+  envValue: string | undefined,
+  localCap: number,
+): Check {
+  const name = 'embed_concurrency';
+  if (!isLocalEndpoint) {
+    return { name, status: 'ok', message: 'Embedding endpoint is not a local inference server — cloud concurrency defaults apply.' };
+  }
+  const parsed = envValue ? parseInt(envValue, 10) : NaN;
+  if (envValue && Number.isFinite(parsed) && parsed > localCap) {
+    return {
+      name,
+      status: 'warn',
+      message:
+        `GBRAIN_EMBED_CONCURRENCY=${parsed} against a local embedding endpoint. ` +
+        `Local inference servers serialize requests, so ${parsed} parallel pages multiply ` +
+        `latency x${parsed} and can exceed the fetch timeout — the embed backfill stalls ` +
+        `with no error. Unset GBRAIN_EMBED_CONCURRENCY (auto-caps at ${localCap}) or set it <= ${localCap}.`,
+    };
+  }
+  return {
+    name,
+    status: 'ok',
+    message: `Local embedding endpoint detected; embed concurrency capped at ${envValue ? parsed : localCap}.`,
+  };
+}
+
+/** Thin gateway/env wrapper over `computeEmbedConcurrencyCheck`. */
+export async function checkEmbedConcurrency(): Promise<Check> {
+  try {
+    const { isLocalEmbeddingEndpoint, LOCAL_EMBED_CONCURRENCY_CAP } = await import('../core/ai/gateway.ts');
+    return computeEmbedConcurrencyCheck(
+      isLocalEmbeddingEndpoint(),
+      process.env.GBRAIN_EMBED_CONCURRENCY,
+      LOCAL_EMBED_CONCURRENCY_CAP,
+    );
+  } catch (err) {
+    return {
+      name: 'embed_concurrency',
+      status: 'ok',
+      message: `Skipped (${err instanceof Error ? err.message : String(err)})`,
+    };
+  }
+}
+
 /** Thin env/engine wrapper over `computePoolBudgetCheck`. */
 export async function checkPoolBudget(_engine: BrainEngine): Promise<Check> {
   try {

src/commands/embed.ts

@@ -1,5 +1,6 @@
 import type { BrainEngine } from '../core/engine.ts';
-import { embedBatch, currentEmbeddingSignature, resolveEmbeddingModelLabel } from '../core/embedding.ts';
+import { embedBatch, currentEmbeddingSignature } from '../core/embedding.ts';
+import { isLocalEmbeddingEndpoint, LOCAL_EMBED_CONCURRENCY_CAP } from '../core/ai/gateway.ts';
 import type { ChunkInput } from '../core/types.ts';
 import { chunkText } from '../core/chunkers/recursive.ts';
 import { createProgress, type ProgressReporter } from '../core/progress.ts';
@@ -176,6 +177,31 @@ export class EmbeddingDimMismatchError extends Error {
   }
 }
 
+/**
+ * #2552: resolve the bulk-embed worker count. Env override or the
+ * cloud-tuned default of 20 — but when the operator did NOT set
+ * GBRAIN_EMBED_CONCURRENCY and the embedding endpoint is a local inference
+ * server (Ollama / llama-server / localhost base URL), cap at
+ * LOCAL_EMBED_CONCURRENCY_CAP: 20 parallel pages against a single-slot
+ * server serialize on the one loaded model, multiply latency x20 past the
+ * fetch timeout, and starve the backfill with no surfaced error. An
+ * explicit env value always wins (`gbrain doctor` warns instead).
+ * Pacing only ever LOWERS concurrency (Codex P2).
+ */
+export function resolveEmbedConcurrency(paceMaxConcurrency?: number): number {
+  const envSet = !!process.env.GBRAIN_EMBED_CONCURRENCY;
+  const base = parseInt(process.env.GBRAIN_EMBED_CONCURRENCY || '20', 10);
+  let resolved = base;
+  if (!envSet && isLocalEmbeddingEndpoint() && base > LOCAL_EMBED_CONCURRENCY_CAP) {
+    resolved = LOCAL_EMBED_CONCURRENCY_CAP;
+    serr(
+      `[embed] local embedding endpoint detected — capping concurrency at ` +
+      `${LOCAL_EMBED_CONCURRENCY_CAP} (set GBRAIN_EMBED_CONCURRENCY to override)`,
+    );
+  }
+  return paceMaxConcurrency ? Math.min(resolved, paceMaxConcurrency) : resolved;
+}
+
 /**
  * Pre-flight check: read the actual schema column dim and compare to the
  * gateway's resolved dim. Throws `EmbeddingDimMismatchError` on mismatch
@@ -581,16 +607,11 @@ async function embedPage(
   for (let j = 0; j < toEmbed.length; j++) {
     embeddingMap.set(toEmbed[j].chunk_index, embeddings[j]);
   }
-  // #1717: label each (re)embedded chunk with the model that actually
-  // produced its vector. Preserved chunks (not re-embedded this pass) keep
-  // their existing model so a mixed-model page isn't relabeled wholesale.
-  const embedModelLabel = resolveEmbeddingModelLabel();
   const updated: ChunkInput[] = chunks.map(c => ({
     chunk_index: c.chunk_index,
     chunk_text: c.chunk_text,
     chunk_source: c.chunk_source,
     embedding: embeddingMap.get(c.chunk_index),
-    model: embeddingMap.has(c.chunk_index) && embedModelLabel ? embedModelLabel : c.model,
     token_count: c.token_count || Math.ceil(c.chunk_text.length / 4),
   }));
 
@@ -682,10 +703,8 @@ async function embedAll(
   // Paced runs lower this to the resolved cap (the real lever vs pooler-slot
   // starvation); unpaced keeps the env/default 20. Codex P2: only ever LOWER —
   // never raise above an operator's existing env cap.
-  const BASE_CONCURRENCY = parseInt(process.env.GBRAIN_EMBED_CONCURRENCY || '20', 10);
-  const CONCURRENCY = staleOpts?.paceMaxConcurrency
-    ? Math.min(BASE_CONCURRENCY, staleOpts.paceMaxConcurrency)
-    : BASE_CONCURRENCY;
+  // #2552: local endpoints auto-cap — see resolveEmbedConcurrency.
+  const CONCURRENCY = resolveEmbedConcurrency(staleOpts?.paceMaxConcurrency);
 
   async function embedOnePage(page: typeof pages[number]) {
     // #1737: bail before doing any work for this page if the run was aborted.
@@ -722,16 +741,12 @@ async function embedAll(
       for (let j = 0; j < toEmbed.length; j++) {
         embeddingMap.set(toEmbed[j].chunk_index, embeddings[j]);
       }
-      // #1717: stamp the resolved embedding model on (re)embedded chunks;
-      // preserve the existing model on chunks left untouched.
-      const embedModelLabel = resolveEmbeddingModelLabel();
       // Preserve ALL chunks, only update embeddings for stale ones
       const updated: ChunkInput[] = chunks.map(c => ({
         chunk_index: c.chunk_index,
         chunk_text: c.chunk_text,
         chunk_source: c.chunk_source,
         embedding: embeddingMap.get(c.chunk_index) ?? undefined,
-        model: embeddingMap.has(c.chunk_index) && embedModelLabel ? embedModelLabel : c.model,
         token_count: c.token_count || Math.ceil(c.chunk_text.length / 4),
       }));
       await observed(pacer, () => engine.upsertChunks(page.slug, updated, pageOpts));
@@ -864,10 +879,8 @@ async function embedAllStale(
   // Paced runs lower concurrency to the resolved cap (E-1: worker count IS the
   // lever on this single pool, no separate permit). Codex P2: pacing only ever
   // LOWERS concurrency — never raise above an operator's existing env cap.
-  const BASE_CONCURRENCY = parseInt(process.env.GBRAIN_EMBED_CONCURRENCY || '20', 10);
-  const CONCURRENCY = staleOpts?.paceMaxConcurrency
-    ? Math.min(BASE_CONCURRENCY, staleOpts.paceMaxConcurrency)
-    : BASE_CONCURRENCY;
+  // #2552: local endpoints auto-cap — see resolveEmbedConcurrency.
+  const CONCURRENCY = resolveEmbedConcurrency(staleOpts?.paceMaxConcurrency);
   const pacer = staleOpts?.pacer ?? createNoopPacer();
 
   // D3 + D3a + D8: wall-clock budget. 30 min default; env override.
@@ -1021,15 +1034,11 @@ async function embedAllStale(
           for (let j = 0; j < stale.length; j++) {
             staleIdxToEmbedding.set(stale[j].chunk_index, embeddings[j]);
           }
-          // #1717: label the re-embedded (stale) chunks with the resolved
-          // model; preserve the existing model on the non-stale chunks.
-          const embedModelLabel = resolveEmbeddingModelLabel();
           const merged: ChunkInput[] = existing.map(c => ({
             chunk_index: c.chunk_index,
             chunk_text: c.chunk_text,
             chunk_source: c.chunk_source,
             embedding: staleIdxToEmbedding.get(c.chunk_index) ?? undefined,
-            model: staleIdxToEmbedding.has(c.chunk_index) && embedModelLabel ? embedModelLabel : c.model,
             token_count: c.token_count || Math.ceil(c.chunk_text.length / 4),
           }));
           await observed(pacer, () => engine.upsertChunks(slug, merged, { sourceId: keySourceId }));

src/core/ai/gateway.ts

@@ -683,6 +683,33 @@ export function getEmbeddingDimensions(): number {
   return requireConfig().embedding_dimensions ?? DEFAULT_EMBEDDING_DIMENSIONS;
 }
 
+/**
+ * #2552: cap for parallel bulk-embed workers against a local inference
+ * server. A single-slot Ollama/llama-server serializes requests, so the
+ * cloud-tuned 20-worker fan-out multiplies latency x20 and blows past the
+ * fetch timeout with no surfaced error (the backfill silently starves).
+ */
+export const LOCAL_EMBED_CONCURRENCY_CAP = 2;
+
+/**
+ * #2552: true when the configured embedding model routes to a local
+ * inference server — the `ollama` / `llama-server` recipes, or any recipe
+ * whose base URL was explicitly pointed at localhost. Bulk callers use this
+ * to pick CPU-safe concurrency defaults; `gbrain doctor` uses it to warn
+ * about an explicit cloud-sized override. Fail-open: unconfigured or
+ * unresolvable gateway → false (cloud behavior, the historical default).
+ */
+export function isLocalEmbeddingEndpoint(): boolean {
+  try {
+    const { recipe } = resolveRecipe(getEmbeddingModel());
+    if (recipe.id === 'ollama' || recipe.id === 'llama-server') return true;
+    const base = requireConfig().base_urls?.[recipe.id] ?? '';
+    return /\/\/(localhost|127\.0\.0\.1|\[::1\])(:|\/|$)/i.test(base);
+  } catch {
+    return false;
+  }
+}
+
 /**
  * v0.28.11: returns the configured multimodal embedding model when set,
  * or undefined if the brain falls back to `embedding_model` for multimodal

src/core/ai/recipes/ollama.ts

@@ -29,9 +29,17 @@ export const ollama: Recipe = {
       trust_custom_dims: true, // #2271: local models carry varied native dims
       cost_per_1m_tokens_usd: 0,
       price_last_verified: '2026-04-20',
-      // Ollama's batch capacity depends on the locally loaded model + the
-      // OLLAMA_NUM_PARALLEL config; no static cap to declare. v0.32 (#779).
-      no_batch_cap: true,
+      // #2552: Ollama's true batch capacity depends on the locally loaded
+      // model + OLLAMA_NUM_PARALLEL, but the previous `no_batch_cap: true`
+      // meant a whole page went out in ONE request — on a CPU-only box that
+      // multiplies latency past the fetch timeout and the backfill starves
+      // with no surfaced error. Ollama doesn't return a recognizable
+      // token-limit error either, so the recursive-halving safety net never
+      // fires; a conservative static pre-split cap is the only guard.
+      // 4096 tokens x 2 chars/token ~= 8K chars per request (code-dense
+      // pages run ~2 chars/token, not the tiktoken-ish 4).
+      max_batch_tokens: 4096,
+      chars_per_token: 2,
     },
   },
   setup_hint: 'Install Ollama from https://ollama.ai, then `ollama pull nomic-embed-text` and `ollama serve`.',

src/core/doctor-categories.ts

@@ -141,6 +141,7 @@ export const OPS_CHECK_NAMES: ReadonlySet<string> = new Set([
   'pgbouncer_prepare',
   'pgvector',
   'pool_budget',
+  'embed_concurrency',
   'progressive_batch_audit_health',
   'queue_health',
   'reranker_health',

src/core/embed-stale.ts

@@ -20,7 +20,6 @@
 import type { BrainEngine } from './engine.ts';
 import type { ChunkInput } from './types.ts';
 import { embedBatchWithBackoff } from '../commands/embed.ts';
-import { resolveEmbeddingModelLabel } from './embedding.ts';
 import { type DbPacer, createNoopPacer, observed } from './db-pacer.ts';
 import { AbortError } from './abort-check.ts';
 
@@ -201,17 +200,11 @@ export async function embedStaleForSource(
         for (let j = 0; j < stale.length; j++) {
           staleIdxToEmbedding.set(stale[j].chunk_index, embeddings[j]);
         }
-        // #1717: label re-embedded chunks with the model that produced the
-        // vector; preserved chunks keep their existing model. Without this,
-        // upsertChunks falls back to DEFAULT_EMBEDDING_MODEL for every chunk
-        // (the same mislabel the embed.ts paths fixed).
-        const embedModelLabel = resolveEmbeddingModelLabel();
         const merged: ChunkInput[] = existing.map((c) => ({
           chunk_index: c.chunk_index,
           chunk_text: c.chunk_text,
           chunk_source: c.chunk_source,
           embedding: staleIdxToEmbedding.get(c.chunk_index) ?? undefined,
-          model: staleIdxToEmbedding.has(c.chunk_index) && embedModelLabel ? embedModelLabel : c.model,
           token_count: c.token_count || Math.ceil(c.chunk_text.length / 4),
           // Carry through per-chunk metadata. upsertChunks writes these as
           // EXCLUDED.<col> (not COALESCE), so omitting them here resets image

src/core/embedding.ts

@@ -113,21 +113,6 @@ export async function embedBatch(
   return results;
 }
 
-/**
- * Resolve the embedding model label (`provider:model`) to stamp onto
- * `content_chunks.model`, so each chunk records the model that actually
- * produced its vector instead of the engine's hardcoded default (#1717).
- * Returns undefined if the gateway is unconfigured; callers then fall back
- * to the chunk's existing model rather than mislabeling it.
- */
-export function resolveEmbeddingModelLabel(): string | undefined {
-  try {
-    return gatewayGetModel();
-  } catch {
-    return undefined;
-  }
-}
-
 /** Currently-configured embedding model (short form without provider prefix). */
 export function getEmbeddingModelName(): string {
   return gatewayGetModel().split(':').slice(1).join(':') || 'text-embedding-3-large';

src/core/import-file.ts

@@ -8,7 +8,7 @@ import { chunkText } from './chunkers/recursive.ts';
 import { chunkCodeText, chunkCodeTextFull, detectCodeLanguage, CHUNKER_VERSION } from './chunkers/code.ts';
 import { findChunkForOffset } from './chunkers/edge-extractor.ts';
 import { extractCodeRefs, imageOfCandidates } from './link-extraction.ts';
-import { embedBatch, embedMultimodal, currentEmbeddingSignature, resolveEmbeddingModelLabel } from './embedding.ts';
+import { embedBatch, embedMultimodal, currentEmbeddingSignature } from './embedding.ts';
 import { slugifyPath, slugifyCodePath, isCodeFilePath } from './sync.ts';
 import type { ChunkInput, PageInput, PageType } from './types.ts';
 import { computeEffectiveDate } from './effective-date.ts';
@@ -716,12 +716,8 @@ export async function importFromContent(
       ? chunks.map((c) => wrapChunkForEmbedding(c.chunk_text, prefix, c.chunk_source))
       : chunks.map((c) => c.chunk_text);
     const embeddings = await embedBatch(wrappedTexts);
-    // #1717: label each chunk with the model that actually produced its
-    // vector, not the engine's hardcoded default.
-    const embedModelLabel = resolveEmbeddingModelLabel();
     for (let i = 0; i < chunks.length; i++) {
       chunks[i].embedding = embeddings[i];
-      if (embedModelLabel) chunks[i].model = embedModelLabel;
       // token_count tracks the wrapped string length so cost reporting
       // reflects what we actually sent to the embedder.
       chunks[i].token_count = Math.ceil(wrappedTexts[i].length / 4);
@@ -1145,10 +1141,7 @@ export async function importCodeFile(
     const matched = existingByKey.get(key);
     if (matched && matched.embedding) {
       // Reuse the existing embedding verbatim. No API call, no cost.
-      // #1717: carry the existing model label along with the reused vector
-      // so the upsert doesn't relabel it with the engine default.
       chunks[i]!.embedding = matched.embedding as Float32Array;
-      chunks[i]!.model = matched.model ?? undefined;
       chunks[i]!.token_count = matched.token_count ?? undefined;
     } else {
       needsEmbedIndexes.push(i);
@@ -1160,12 +1153,9 @@ export async function importCodeFile(
     try {
       const textsToEmbed = needsEmbedIndexes.map((i) => chunks[i]!.chunk_text);
       const embeddings = await embedBatch(textsToEmbed);
-      // #1717: stamp the model that produced these vectors.
-      const embedModelLabel = resolveEmbeddingModelLabel();
       for (let j = 0; j < needsEmbedIndexes.length; j++) {
         const i = needsEmbedIndexes[j]!;
         chunks[i]!.embedding = embeddings[j]!;
-        if (embedModelLabel) chunks[i]!.model = embedModelLabel;
         chunks[i]!.token_count = Math.ceil(chunks[i]!.chunk_text.length / 4);
       }
     } catch (e: unknown) {

src/core/pglite-engine.ts

@@ -56,7 +56,7 @@ import { GBrainError, PAGE_SORT_SQL, ENRICH_ORDER_SQL } from './types.ts';
 import { finalizeLastSeen } from './chronicle/last-seen.ts';
 import { computeAnomaliesFromBuckets } from './cycle/anomaly.ts';
 import { resolveBoostMap, resolveHardExcludes } from './search/source-boost.ts';
-import { buildSourceFactorCase, buildHardExcludeClause, buildVisibilityClause, buildRecencyComponentSql, buildBestPerPagePoolCte, buildOrFallbackWebsearchQuery } from './search/sql-ranking.ts';
+import { buildSourceFactorCase, buildHardExcludeClause, buildVisibilityClause, buildRecencyComponentSql, buildBestPerPagePoolCte, buildOrFallbackWebsearchQuery, buildWebsearchQueryExpr } from './search/sql-ranking.ts';
 import {
   normalizeEngineColumn,
   buildVectorCastFragment,
@@ -1591,6 +1591,8 @@ export class PGLiteEngine implements BrainEngine {
     }
 
     // v0.20.0 Cathedral II Layer 10 C1/C2: language + symbol-kind filters.
+    // #2380: slash-bearing queries match both the split-word and literal
+    // slash forms — see buildWebsearchQueryExpr in ./search/sql-ranking.ts.
     const params: unknown[] = [query, innerLimit, limit, offset];
     let extraFilter = '';
     if (opts?.language) {
@@ -1630,6 +1632,7 @@ export class PGLiteEngine implements BrainEngine {
     // FTS config name (e.g. 'english', 'pt_br'). Validated by getFtsLanguage()
     // — safe to interpolate into raw SQL.
     const ftsLang = getFtsLanguage();
+    const ftsQueryExpr = buildWebsearchQueryExpr(ftsLang, '$1', query);
 
     const keywordSql =
       `WITH ranked AS (
@@ -1637,14 +1640,14 @@ export class PGLiteEngine implements BrainEngine {
            p.slug, p.id as page_id, p.title, p.type, p.source_id,
            p.effective_date, p.effective_date_source,
            cc.id as chunk_id, cc.chunk_index, cc.chunk_text, cc.chunk_source,
-           ts_rank(cc.search_vector, websearch_to_tsquery('${ftsLang}', $1)) * ${sourceFactorCase} AS score,
+           ts_rank(cc.search_vector, ${ftsQueryExpr}) * ${sourceFactorCase} AS score,
            CASE WHEN p.updated_at < (
              SELECT MAX(te.created_at) FROM timeline_entries te WHERE te.page_id = p.id
            ) THEN true ELSE false END AS stale
          FROM content_chunks cc
          JOIN pages p ON p.id = cc.page_id
          JOIN sources s ON s.id = p.source_id
-         WHERE cc.search_vector @@ websearch_to_tsquery('${ftsLang}', $1) ${detailFilter}${extraFilter} ${hardExcludeClause} ${visibilityClause}
+         WHERE cc.search_vector @@ ${ftsQueryExpr} ${detailFilter}${extraFilter} ${hardExcludeClause} ${visibilityClause}
            -- v0.27.1: hide image rows from default text-keyword search so
            -- OCR text doesn't drown text-page hits. Image-similarity queries
            -- run a separate vector path on embedding_image.
@@ -1712,7 +1715,10 @@ export class PGLiteEngine implements BrainEngine {
     // FTS config name (e.g. 'english', 'pt_br'). Validated by getFtsLanguage()
     // — safe to interpolate into raw SQL.
     const ftsLang = getFtsLanguage();
+    const ftsQueryExpr = buildWebsearchQueryExpr(ftsLang, '$1', query);
 
+    // #2380: slash-bearing queries match both the split-word and literal
+    // slash forms — see buildWebsearchQueryExpr in ./search/sql-ranking.ts.
     const params: unknown[] = [query, limit, offset];
     let extraFilter = '';
     if (opts?.type) {
@@ -1760,7 +1766,7 @@ export class PGLiteEngine implements BrainEngine {
          COALESCE(rep.chunk_index, 0) as chunk_index,
          COALESCE(rep.chunk_text, '') as chunk_text,
          COALESCE(rep.chunk_source, 'compiled_truth') as chunk_source,
-         ts_rank_cd(p.search_vector, websearch_to_tsquery('${ftsLang}', $1)) * ${sourceFactorCase} AS score,
+         ts_rank_cd(p.search_vector, ${ftsQueryExpr}) * ${sourceFactorCase} AS score,
          CASE WHEN p.updated_at < (
            SELECT MAX(te.created_at) FROM timeline_entries te WHERE te.page_id = p.id
          ) THEN true ELSE false END AS stale
@@ -1775,7 +1781,7 @@ export class PGLiteEngine implements BrainEngine {
          ORDER BY (cc.chunk_source = 'compiled_truth') DESC, cc.chunk_index ASC
          LIMIT 1
        ) rep ON true
-       WHERE p.search_vector @@ websearch_to_tsquery('${ftsLang}', $1)
+       WHERE p.search_vector @@ ${ftsQueryExpr}
          ${extraFilter} ${hardExcludeClause} ${visibilityClause}
        ORDER BY score DESC, p.id ASC
        LIMIT $2 OFFSET $3`;
@@ -1962,6 +1968,8 @@ export class PGLiteEngine implements BrainEngine {
       });
     }
 
+    // #2380: slash-bearing queries match both the split-word and literal
+    // slash forms — see buildWebsearchQueryExpr in ./search/sql-ranking.ts.
     const params: unknown[] = [query, limit, offset];
     let extraFilter = '';
     if (opts?.language) {
@@ -1996,20 +2004,21 @@ export class PGLiteEngine implements BrainEngine {
     // FTS config name (e.g. 'english', 'pt_br'). Validated by getFtsLanguage()
     // — safe to interpolate into raw SQL.
     const ftsLang = getFtsLanguage();
+    const ftsQueryExpr = buildWebsearchQueryExpr(ftsLang, '$1', query);
 
     const { rows } = await this.db.query(
       `SELECT
          p.slug, p.id as page_id, p.title, p.type, p.source_id,
          p.effective_date, p.effective_date_source,
          cc.id as chunk_id, cc.chunk_index, cc.chunk_text, cc.chunk_source,
-         ts_rank(cc.search_vector, websearch_to_tsquery('${ftsLang}', $1)) * ${sourceFactorCase} AS score,
+         ts_rank(cc.search_vector, ${ftsQueryExpr}) * ${sourceFactorCase} AS score,
          CASE WHEN p.updated_at < (
            SELECT MAX(te.created_at) FROM timeline_entries te WHERE te.page_id = p.id
          ) THEN true ELSE false END AS stale
        FROM content_chunks cc
        JOIN pages p ON p.id = cc.page_id
        JOIN sources s ON s.id = p.source_id
-       WHERE cc.search_vector @@ websearch_to_tsquery('${ftsLang}', $1) ${detailFilter}${extraFilter} ${hardExcludeClause} ${visibilityClause}
+       WHERE cc.search_vector @@ ${ftsQueryExpr} ${detailFilter}${extraFilter} ${hardExcludeClause} ${visibilityClause}
        ORDER BY score DESC
        LIMIT $2 OFFSET $3`,
       params

src/core/postgres-engine.ts

@@ -64,7 +64,7 @@ import { ConnectionManager } from './connection-manager.ts';
 import { logConnectionEvent } from './connection-audit.ts';
 import { validateSlug, contentHash, rowToPage, rowToStalePage, rowToChunk, rowToSearchResult, parseEmbedding, tryParseEmbedding, takeRowToTake, takeHitRowToHit, isUndefinedTableError, warnOncePerProcess } from './utils.ts';
 import { resolveBoostMap, resolveHardExcludes } from './search/source-boost.ts';
-import { buildSourceFactorCase, buildHardExcludeClause, buildVisibilityClause, buildRecencyComponentSql, buildBestPerPagePoolCte, buildOrFallbackWebsearchQuery } from './search/sql-ranking.ts';
+import { buildSourceFactorCase, buildHardExcludeClause, buildVisibilityClause, buildRecencyComponentSql, buildBestPerPagePoolCte, buildOrFallbackWebsearchQuery, buildWebsearchQueryExpr } from './search/sql-ranking.ts';
 import { DEFAULT_EMBEDDING_MODEL, DEFAULT_EMBEDDING_DIMENSIONS } from './ai/defaults.ts';
 import { DELETE_BATCH_SIZE } from './engine-constants.ts';
 
@@ -1691,6 +1691,8 @@ export class PostgresEngine implements BrainEngine {
     const hardExcludePrefixes = resolveHardExcludes(opts?.exclude_slug_prefixes, opts?.include_slug_prefixes);
     const hardExcludeClause = buildHardExcludeClause('p.slug', hardExcludePrefixes);
 
+    // #2380: slash-bearing queries match both the split-word and literal
+    // slash forms — see buildWebsearchQueryExpr in ./search/sql-ranking.ts.
     const params: unknown[] = [query];
     let typeClause = '';
     if (type) {
@@ -1761,6 +1763,7 @@ export class PostgresEngine implements BrainEngine {
     // FTS config name (e.g. 'english', 'pt_br'). Validated by getFtsLanguage()
     // — safe to interpolate into raw SQL.
     const ftsLang = getFtsLanguage();
+    const ftsQueryExpr = buildWebsearchQueryExpr(ftsLang, '$1', query);
 
     const rawQuery = `
       WITH ranked_chunks AS (
@@ -1768,11 +1771,11 @@ export class PostgresEngine implements BrainEngine {
           p.slug, p.id as page_id, p.title, p.type, p.source_id,
           p.effective_date, p.effective_date_source,
           cc.id as chunk_id, cc.chunk_index, cc.chunk_text, cc.chunk_source,
-          ts_rank(cc.search_vector, websearch_to_tsquery('${ftsLang}', $1)) * ${sourceFactorCase} AS score
+          ts_rank(cc.search_vector, ${ftsQueryExpr}) * ${sourceFactorCase} AS score
         FROM content_chunks cc
         JOIN pages p ON p.id = cc.page_id
         JOIN sources s ON s.id = p.source_id
-        WHERE cc.search_vector @@ websearch_to_tsquery('${ftsLang}', $1)
+        WHERE cc.search_vector @@ ${ftsQueryExpr}
           ${typeClause}
           ${typesClause}
           ${excludeSlugsClause}
@@ -1863,7 +1866,10 @@ export class PostgresEngine implements BrainEngine {
     // FTS config name (e.g. 'english', 'pt_br'). Validated by getFtsLanguage()
     // — safe to interpolate into raw SQL.
     const ftsLang = getFtsLanguage();
+    const ftsQueryExpr = buildWebsearchQueryExpr(ftsLang, '$1', query);
 
+    // #2380: slash-bearing queries match both the split-word and literal
+    // slash forms — see buildWebsearchQueryExpr in ./search/sql-ranking.ts.
     const params: unknown[] = [query];
     let typeClause = '';
     if (opts?.type) {
@@ -1923,7 +1929,7 @@ export class PostgresEngine implements BrainEngine {
         COALESCE(rep.chunk_index, 0) as chunk_index,
         COALESCE(rep.chunk_text, '') as chunk_text,
         COALESCE(rep.chunk_source, 'compiled_truth') as chunk_source,
-        ts_rank_cd(p.search_vector, websearch_to_tsquery('${ftsLang}', $1)) * ${sourceFactorCase} AS score,
+        ts_rank_cd(p.search_vector, ${ftsQueryExpr}) * ${sourceFactorCase} AS score,
         false AS stale
       FROM pages p
       JOIN sources s ON s.id = p.source_id
@@ -1936,7 +1942,7 @@ export class PostgresEngine implements BrainEngine {
         ORDER BY (cc.chunk_source = 'compiled_truth') DESC, cc.chunk_index ASC
         LIMIT 1
       ) rep ON true
-      WHERE p.search_vector @@ websearch_to_tsquery('${ftsLang}', $1)
+      WHERE p.search_vector @@ ${ftsQueryExpr}
         ${typeClause}
         ${typesClause}
         ${excludeSlugsClause}
@@ -2000,6 +2006,8 @@ export class PostgresEngine implements BrainEngine {
     const hardExcludePrefixes = resolveHardExcludes(opts?.exclude_slug_prefixes, opts?.include_slug_prefixes);
     const hardExcludeClause = buildHardExcludeClause('p.slug', hardExcludePrefixes);
 
+    // #2380: slash-bearing queries match both the split-word and literal
+    // slash forms — see buildWebsearchQueryExpr in ./search/sql-ranking.ts.
     const params: unknown[] = [query];
     let typeClause = '';
     if (type) {
@@ -2060,18 +2068,19 @@ export class PostgresEngine implements BrainEngine {
     // FTS config name (e.g. 'english', 'pt_br'). Validated by getFtsLanguage()
     // — safe to interpolate into raw SQL.
     const ftsLang = getFtsLanguage();
+    const ftsQueryExpr = buildWebsearchQueryExpr(ftsLang, '$1', query);
 
     const rawQuery = `
       SELECT
         p.slug, p.id as page_id, p.title, p.type, p.source_id,
         p.effective_date, p.effective_date_source,
         cc.id as chunk_id, cc.chunk_index, cc.chunk_text, cc.chunk_source,
-        ts_rank(cc.search_vector, websearch_to_tsquery('${ftsLang}', $1)) * ${sourceFactorCase} AS score,
+        ts_rank(cc.search_vector, ${ftsQueryExpr}) * ${sourceFactorCase} AS score,
         false AS stale
       FROM content_chunks cc
       JOIN pages p ON p.id = cc.page_id
       JOIN sources s ON s.id = p.source_id
-      WHERE cc.search_vector @@ websearch_to_tsquery('${ftsLang}', $1)
+      WHERE cc.search_vector @@ ${ftsQueryExpr}
         ${typeClause}
         ${typesClause}
         ${excludeSlugsClause}

src/core/search/sql-ranking.ts

@@ -251,6 +251,28 @@ export function buildOrFallbackWebsearchQuery(query: string): string | null {
   return tokens.join(' OR ');
 }
 
+/**
+ * #2380: FTS query expression for slash-bearing queries. Postgres' default
+ * text-search parser classifies `foo/bar` as a single `file`-alias lexeme —
+ * on BOTH the query side and the index side. So a raw `foo/bar` query only
+ * matched documents carrying the identical joined lexeme (literal paths),
+ * and a slash-split query only matches documents whose text had the words
+ * separated. Neither form alone covers both document shapes; OR the two
+ * parses so a slash query matches prose ("foo and bar", stemmed, AND
+ * semantics) AND literal slash forms ("src/core/x.ts") alike.
+ *
+ * Slash-free queries return the plain single-parse expression — byte-
+ * identical SQL and identical ts_rank to the historical behavior.
+ *
+ * `ftsLang` is validated by getFtsLanguage() (safe to interpolate);
+ * `param` is a `$N` placeholder, never user text.
+ */
+export function buildWebsearchQueryExpr(ftsLang: string, param: string, query: string): string {
+  const plain = `websearch_to_tsquery('${ftsLang}', ${param})`;
+  if (!query.includes('/')) return plain;
+  return `(websearch_to_tsquery('${ftsLang}', translate(${param}, '/', ' ')) || ${plain})`;
+}
+
 // ============================================================
 // v0.29.1 — Recency component SQL builder
 // ============================================================

test/ai/no-batch-cap-suppression.serial.test.ts

@@ -28,8 +28,8 @@ describe('v0.32 #779: no_batch_cap suppresses the missing-max_batch_tokens warni
     resetGateway();
   });
 
-  test('Ollama, LiteLLM, llama-server all declare no_batch_cap: true', () => {
-    for (const id of ['ollama', 'litellm', 'llama-server']) {
+  test('LiteLLM and llama-server declare no_batch_cap: true', () => {
+    for (const id of ['litellm', 'llama-server']) {
       const r = getRecipe(id);
       expect(r, `${id} not registered`).toBeDefined();
       expect(
@@ -39,6 +39,18 @@ describe('v0.32 #779: no_batch_cap suppresses the missing-max_batch_tokens warni
     }
   });
 
+  test('#2552: Ollama declares a conservative static batch cap, not no_batch_cap', () => {
+    // A CPU-only Ollama box wedges when a whole page ships in one request;
+    // Ollama never returns a token-limit error so the recursive-halving
+    // safety net can't fire. The pre-split cap is the only guard.
+    const r = getRecipe('ollama');
+    expect(r).toBeDefined();
+    const e = r!.touchpoints.embedding!;
+    expect(e.no_batch_cap).toBeUndefined();
+    expect(e.max_batch_tokens).toBe(4096);
+    expect(e.chars_per_token).toBe(2);
+  });
+
   test('configureGateway does NOT warn for ollama/litellm/llama-server', () => {
     warnSpy.mockClear();
     resetGateway();

test/embed-local-endpoint.serial.test.ts

@@ -0,0 +1,118 @@
+/**
+ * #2552: cloud-tuned embedding defaults silently wedge CPU-only local
+ * endpoints (Ollama). Three-part fix under test:
+ *
+ *  1. `isLocalEmbeddingEndpoint()` — gateway helper detecting local
+ *     inference servers (ollama / llama-server recipes, localhost base URL).
+ *  2. `resolveEmbedConcurrency()` — embed auto-caps the 20-worker fan-out
+ *     at LOCAL_EMBED_CONCURRENCY_CAP for local endpoints unless the
+ *     operator set GBRAIN_EMBED_CONCURRENCY explicitly.
+ *  3. `computeEmbedConcurrencyCheck()` — doctor warns when an explicit env
+ *     override fans out against a local endpoint.
+ *
+ * Serial: mutates process.env and the module-global gateway config.
+ */
+
+import { afterAll, afterEach, describe, expect, test } from 'bun:test';
+import {
+  configureGateway,
+  resetGateway,
+  isLocalEmbeddingEndpoint,
+  LOCAL_EMBED_CONCURRENCY_CAP,
+} from '../src/core/ai/gateway.ts';
+import { resolveEmbedConcurrency } from '../src/commands/embed.ts';
+import { computeEmbedConcurrencyCheck } from '../src/commands/doctor.ts';
+
+const SAVED_ENV = process.env.GBRAIN_EMBED_CONCURRENCY;
+
+afterEach(() => {
+  resetGateway();
+  if (SAVED_ENV === undefined) delete process.env.GBRAIN_EMBED_CONCURRENCY;
+  else process.env.GBRAIN_EMBED_CONCURRENCY = SAVED_ENV;
+});
+
+afterAll(() => {
+  resetGateway();
+});
+
+describe('#2552 isLocalEmbeddingEndpoint', () => {
+  test('false when the gateway is not configured (fail-open to cloud behavior)', () => {
+    resetGateway();
+    expect(isLocalEmbeddingEndpoint()).toBe(false);
+  });
+
+  test('true for the ollama recipe', () => {
+    configureGateway({ embedding_model: 'ollama:nomic-embed-text', env: {} });
+    expect(isLocalEmbeddingEndpoint()).toBe(true);
+  });
+
+  test('true for the llama-server recipe', () => {
+    configureGateway({ embedding_model: 'llama-server:my-gguf', env: {} });
+    expect(isLocalEmbeddingEndpoint()).toBe(true);
+  });
+
+  test('false for a cloud recipe', () => {
+    configureGateway({
+      embedding_model: 'openai:text-embedding-3-small',
+      env: { OPENAI_API_KEY: 'fake' },
+    });
+    expect(isLocalEmbeddingEndpoint()).toBe(false);
+  });
+
+  test('true when a cloud recipe base URL is explicitly pointed at localhost', () => {
+    configureGateway({
+      embedding_model: 'openai:text-embedding-3-small',
+      env: { OPENAI_API_KEY: 'fake' },
+      base_urls: { openai: 'http://localhost:8080/v1' },
+    });
+    expect(isLocalEmbeddingEndpoint()).toBe(true);
+  });
+});
+
+describe('#2552 resolveEmbedConcurrency', () => {
+  test('caps at LOCAL_EMBED_CONCURRENCY_CAP for a local endpoint when env is unset', () => {
+    delete process.env.GBRAIN_EMBED_CONCURRENCY;
+    configureGateway({ embedding_model: 'ollama:nomic-embed-text', env: {} });
+    expect(resolveEmbedConcurrency()).toBe(LOCAL_EMBED_CONCURRENCY_CAP);
+  });
+
+  test('explicit env override always wins, even against a local endpoint', () => {
+    process.env.GBRAIN_EMBED_CONCURRENCY = '10';
+    configureGateway({ embedding_model: 'ollama:nomic-embed-text', env: {} });
+    expect(resolveEmbedConcurrency()).toBe(10);
+  });
+
+  test('cloud endpoints keep the historical default of 20', () => {
+    delete process.env.GBRAIN_EMBED_CONCURRENCY;
+    configureGateway({ env: { OPENAI_API_KEY: 'fake' } });
+    expect(resolveEmbedConcurrency()).toBe(20);
+  });
+
+  test('pacing only ever lowers concurrency', () => {
+    delete process.env.GBRAIN_EMBED_CONCURRENCY;
+    configureGateway({ embedding_model: 'ollama:nomic-embed-text', env: {} });
+    expect(resolveEmbedConcurrency(1)).toBe(1);
+    expect(resolveEmbedConcurrency(16)).toBe(LOCAL_EMBED_CONCURRENCY_CAP);
+  });
+});
+
+describe('#2552 computeEmbedConcurrencyCheck (doctor)', () => {
+  test('ok for non-local endpoints', () => {
+    expect(computeEmbedConcurrencyCheck(false, '20', 2).status).toBe('ok');
+  });
+
+  test('warn when an explicit override exceeds the local cap', () => {
+    const check = computeEmbedConcurrencyCheck(true, '20', 2);
+    expect(check.status).toBe('warn');
+    expect(check.message).toContain('GBRAIN_EMBED_CONCURRENCY=20');
+  });
+
+  test('ok when env is unset against a local endpoint (auto-cap applies)', () => {
+    expect(computeEmbedConcurrencyCheck(true, undefined, 2).status).toBe('ok');
+  });
+
+  test('ok when the override is at or under the cap', () => {
+    expect(computeEmbedConcurrencyCheck(true, '2', 2).status).toBe('ok');
+    expect(computeEmbedConcurrencyCheck(true, '1', 2).status).toBe('ok');
+  });
+});

test/embed-stale.serial.test.ts

@@ -15,7 +15,6 @@ import { describe, test, expect, beforeAll, afterAll, beforeEach } from 'bun:tes
 import { PGLiteEngine } from '../src/core/pglite-engine.ts';
 import { resetPgliteState } from './helpers/reset-pglite.ts';
 import { embedStaleForSource } from '../src/core/embed-stale.ts';
-import { configureGateway, resetGateway } from '../src/core/ai/gateway.ts';
 import type { ChunkInput } from '../src/core/types.ts';
 
 let engine: PGLiteEngine;
@@ -277,49 +276,4 @@ describe('embedStaleForSource', () => {
     // The stale text row actually got its embedding.
     expect(txtRow.embedded_at).not.toBeNull();
   });
-
-  // #1717: the backfill path must label re-embedded chunks with the model
-  // that produced the vector, and preserve the existing label on chunks it
-  // did not touch (before the fix, both were reset to the engine default).
-  test('labels re-embedded chunks with the gateway model, preserves untouched labels (#1717)', async () => {
-    configureGateway({
-      embedding_model: 'openai:text-embedding-3-large',
-      env: { OPENAI_API_KEY: 'sk-test-embed-stale-1717' },
-    });
-    try {
-      await engine.putPage('notes/model-label', {
-        type: 'note',
-        title: 'model-label',
-        compiled_truth: '# model-label\n\nseeded',
-      });
-      await engine.upsertChunks('notes/model-label', [
-        {
-          chunk_index: 0,
-          chunk_text: 'already embedded elsewhere',
-          chunk_source: 'compiled_truth',
-          embedding: new Float32Array(1536).fill(0.01),
-          model: 'voyage:voyage-3',
-          token_count: 4,
-        },
-        {
-          chunk_index: 1,
-          chunk_text: 'stale chunk needing embed',
-          chunk_source: 'compiled_truth',
-          token_count: 5,
-          embedding: undefined, // stale
-        },
-      ]);
-
-      const result = await embedStaleForSource(engine, 'default', { embedFn: fakeEmbedFn });
-      expect(result.embedded).toBe(1);
-
-      const after = await engine.getChunks('notes/model-label');
-      const preserved = after.find((c) => c.chunk_index === 0)!;
-      const reembedded = after.find((c) => c.chunk_index === 1)!;
-      expect(reembedded.model).toBe('openai:text-embedding-3-large');
-      expect(preserved.model).toBe('voyage:voyage-3');
-    } finally {
-      resetGateway();
-    }
-  });
 });

test/embed.serial.test.ts

@@ -37,8 +37,6 @@ mock.module('../src/core/embedding.ts', () => ({
   // setPageEmbeddingSignature / invalidateStaleSignatureEmbeddings resolve to
   // null via the Proxy default, so the signature value is inert here.
   currentEmbeddingSignature: () => 'test:model:1536',
-  // #1717: embed paths stamp this label on (re)embedded chunks.
-  resolveEmbeddingModelLabel: () => 'openai:text-embedding-3-large',
 }));
 
 // Import AFTER mocking.
@@ -805,34 +803,3 @@ describe('embedAllStale --source threading (D7)', () => {
     expect((firstCallOpts as { sourceId?: string }).sourceId).toBe('media-corpus');
   });
 });
-
-// #1717: content_chunks.model must record the model that actually produced
-// each vector, not the gateway/engine default.
-describe('content_chunks.model labeling (#1717)', () => {
-  test('stamps the resolved embedding model on re-embedded chunks, preserves it on untouched chunks', async () => {
-    let upserted: any[] | undefined;
-    // Chunk 0 is stale (no embedded_at) → gets re-embedded this pass.
-    // Chunk 1 is already embedded with a DIFFERENT model → must be preserved,
-    // not relabeled to the current model.
-    const chunks = [
-      { chunk_index: 0, chunk_text: 'a', chunk_source: 'compiled_truth', embedded_at: null, model: 'zeroentropyai:zembed-1', token_count: 1 },
-      { chunk_index: 1, chunk_text: 'b', chunk_source: 'compiled_truth', embedded_at: '2026-01-01', embedding: new Float32Array(1536), model: 'voyage:voyage-3', token_count: 1 },
-    ];
-    const engine = mockEngine({
-      getPage: async () => ({ slug: 'notes/x', compiled_truth: 'a', timeline: '', source_id: 'default' }),
-      getChunks: async () => chunks,
-      upsertChunks: async (_slug: string, c: any[]) => { upserted = c; },
-      setPageEmbeddingSignature: async () => null,
-    });
-
-    await runEmbedCore(engine, { slugs: ['notes/x'] });
-
-    expect(upserted).toBeDefined();
-    const byIdx = Object.fromEntries(upserted!.map(c => [c.chunk_index, c]));
-    // Re-embedded chunk carries the model that produced its vector (was
-    // mislabeled with the default before the fix).
-    expect(byIdx[0].model).toBe('openai:text-embedding-3-large');
-    // Untouched chunk keeps its original model — no wholesale relabel.
-    expect(byIdx[1].model).toBe('voyage:voyage-3');
-  });
-});

test/import-signature-stamp.serial.test.ts

@@ -73,21 +73,4 @@ describe('importFromContent embedding_signature stamping (F1)', () => {
     await importFromContent(engine, 'concepts/unstamped', '# Unstamped\n\nbody content.', { noEmbed: true });
     expect(await signatureOf('concepts/unstamped')).toBeNull();
   });
-
-  // #1717: content_chunks.model must record the model that produced the
-  // vector (the configured gateway model), not the engine's hardcoded
-  // default. The gateway here is configured to openai:text-embedding-3-large,
-  // which differs from DEFAULT_EMBEDDING_MODEL — so this fails without the
-  // import-path model stamping.
-  test('inline embed labels content_chunks.model with the configured model (#1717)', async () => {
-    await importFromContent(engine, 'concepts/labeled', '# Labeled\n\nsome body content to chunk and embed.', {});
-    const rows = await engine.executeRaw<{ model: string }>(
-      `SELECT cc.model FROM content_chunks cc
-        JOIN pages p ON p.id = cc.page_id
-       WHERE p.slug = $1 AND p.source_id = 'default'`,
-      ['concepts/labeled'],
-    );
-    expect(rows.length).toBeGreaterThan(0);
-    for (const r of rows) expect(r.model).toBe('openai:text-embedding-3-large');
-  });
 });

test/pglite-engine.test.ts

@@ -216,6 +216,67 @@ describe('PGLiteEngine: Search', () => {
     expect(results.length).toBe(0);
   });
 
+  // Regression (#2380): queries containing `/` used to bypass FTS AND
+  // semantics. Postgres' default text-search parser classifies `foo/bar` as
+  // a `file`-alias token mapped to the `simple` dictionary, so it became a
+  // single un-stemmed lexeme `'foo/bar'` that never matches indexed text —
+  // the primary FTS pass returned 0 and the OR fallback took over, matching
+  // pages that contain EITHER term. searchKeyword/searchTitles now normalize
+  // `/` to whitespace before websearch_to_tsquery parses, so the primary
+  // AND pass matches directly.
+  test('searchKeyword: slash query matches with AND semantics, not OR fallback', async () => {
+    // Decoy shares only ONE of the two query terms ('enterprise').
+    await engine.putPage('concepts/enterprise-pricing', {
+      type: 'concept', title: 'Widget Pricing',
+      compiled_truth: 'Enterprise pricing for widgets.',
+    });
+    await engine.upsertChunks('concepts/enterprise-pricing', [
+      { chunk_index: 0, chunk_text: 'Enterprise pricing for widgets', chunk_source: 'compiled_truth' },
+    ]);
+
+    // Both terms co-occur only in the novamind chunk. Pre-fix this returned
+    // BOTH pages (primary pass zero-hit → OR fallback); post-fix the primary
+    // AND pass returns exactly the co-occurrence page.
+    const results = await engine.searchKeyword('NovaMind/enterprise');
+    expect(results.length).toBe(1);
+    expect(results[0].slug).toBe('companies/novamind');
+  });
+
+  test('searchTitles: slash query matches with AND semantics, not OR fallback', async () => {
+    await engine.putPage('companies/novamind-enterprise', {
+      type: 'company', title: 'NovaMind Enterprise Platform',
+      compiled_truth: 'Placeholder body.',
+    });
+    await engine.putPage('guides/enterprise-sales', {
+      type: 'concept', title: 'Enterprise Sales Guide',
+      compiled_truth: 'Placeholder body.',
+    });
+
+    // Pre-fix: `NovaMind/Enterprise` parsed as one file-alias lexeme → the
+    // primary title pass returned 0 and the OR fallback matched BOTH titles.
+    const results = await engine.searchTitles('NovaMind/Enterprise');
+    expect(results.length).toBe(1);
+    expect(results[0].slug).toBe('companies/novamind-enterprise');
+  });
+
+  test('searchKeyword: slash query still matches the literal slash form (file paths)', async () => {
+    // The INDEX side also emits the joined file-alias lexeme for literal
+    // `foo/bar` text, so a query normalized to split words alone would go
+    // blind to documents containing the literal slash form (paths, URLs).
+    // buildWebsearchQueryExpr ORs both parses; this pins the raw arm.
+    await engine.putPage('runbooks/widget-deploy', {
+      type: 'concept', title: 'Widget Deploy Runbook',
+      compiled_truth: 'Runbook for the acme/widget deployment pipeline.',
+    });
+    await engine.upsertChunks('runbooks/widget-deploy', [
+      { chunk_index: 0, chunk_text: 'Runbook for the acme/widget deployment pipeline', chunk_source: 'compiled_truth' },
+    ]);
+
+    const results = await engine.searchKeyword('acme/widget');
+    expect(results.length).toBe(1);
+    expect(results[0].slug).toBe('runbooks/widget-deploy');
+  });
+
   test('tsvector trigger populates search_vector on insert', async () => {
     // Verify the PL/pgSQL trigger fires and content_chunks.search_vector is
     // populated from chunk_text. v0.20.0 Cathedral II Layer 3 moved FTS from