fix(search): honor per-call source_id/all_sources in the search op — the #1484 hint's advised retry was silently ignored

The zero-hit CLI hint tells users to retry with --source-id __all__, which
works for `query` (routes through resolveRequestedScope) but the `search`
op had no source_id/all_sources params: the flag parsed into params, nothing
consumed it, and the retry re-ran the same single-source search — an
invisible false negative whose params.source_id also suppressed the hint on
the retry. search now resolves per-call scope through the same canonical
fail-closed resolver as query (local __all__ spans the brain; remote
collapses to the caller's grant; out-of-grant source_id is denied).

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

src/cli.ts

@@ -466,6 +466,11 @@ async function main() {
     const result = JSON.parse(JSON.stringify(rawResult, bigintToStringReplacer));
     const output = formatResult(op.name, result);
     if (output) process.stdout.write(output);
+    // #1484 — invisible-miss hint: a bare query/search that hit zero results
+    // on a multi-source brain tells the user (stderr) which source it
+    // actually searched and how to widen the scope.
+    const hint = await sourceScopeHint(op.name, params, ctx.sourceId, engine, result);
+    if (hint) console.error(hint);
   } catch (e: unknown) {
     // v0.42.20.0 (codex D4): on error, set exitCode + return so the `finally`
     // STILL runs (drains every background-work sink + disconnects). A bare
@@ -837,6 +842,44 @@ async function makeContext(engine: BrainEngine, params: Record<string, unknown>)
   };
 }
 
+/**
+ * #1484 — a bare `gbrain query`/`search` silently scopes to the resolved
+ * source (usually 'default'); on a multi-source brain a zero-hit run looks
+ * identical to "the brain doesn't know this" even when the answer lives in
+ * another source. Returns a stderr hint when (a) the op is query/search,
+ * (b) it returned zero results, (c) the caller did NOT scope explicitly
+ * (--source / --source-id / --all-sources), and (d) the brain has >1
+ * registered source. Best-effort: any lookup failure returns null.
+ *
+ * Exported for tests (same import-safety contract as formatResult).
+ */
+export async function sourceScopeHint(
+  opName: string,
+  params: Record<string, unknown>,
+  sourceId: string,
+  engine: BrainEngine,
+  result: unknown,
+): Promise<string | null> {
+  if (opName !== 'query' && opName !== 'search') return null;
+  if (!Array.isArray(result) || result.length > 0) return null;
+  // Explicit scoping (flag tier) = user intent; don't second-guess it.
+  if (params.source || params.source_id || params.all_sources) return null;
+  if (sourceId === '__all__') return null;
+  try {
+    const rows = await engine.executeRaw<{ n: number }>(
+      `SELECT count(*)::int AS n FROM sources`,
+    );
+    const n = Number(rows[0]?.n ?? 0);
+    if (n <= 1) return null;
+    return (
+      `Hint: this brain has ${n} sources; you searched only "${sourceId}". ` +
+      `Retry with --source-id __all__ (all sources) or --source-id <id>.`
+    );
+  } catch {
+    return null; // hint is best-effort; never fail the query over it
+  }
+}
+
 // Exported for tests (same import-safety contract as cliAliases/printOpHelp).
 export function formatResult(opName: string, result: unknown): string {
   switch (opName) {

src/commands/doctor.ts

@@ -820,10 +820,6 @@ 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.
@@ -3819,61 +3815,6 @@ 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

@@ -1,6 +1,5 @@
 import type { BrainEngine } from '../core/engine.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';
@@ -177,31 +176,6 @@ 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
@@ -703,8 +677,10 @@ 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.
-  // #2552: local endpoints auto-cap — see resolveEmbedConcurrency.
-  const CONCURRENCY = resolveEmbedConcurrency(staleOpts?.paceMaxConcurrency);
+  const BASE_CONCURRENCY = parseInt(process.env.GBRAIN_EMBED_CONCURRENCY || '20', 10);
+  const CONCURRENCY = staleOpts?.paceMaxConcurrency
+    ? Math.min(BASE_CONCURRENCY, staleOpts.paceMaxConcurrency)
+    : BASE_CONCURRENCY;
 
   async function embedOnePage(page: typeof pages[number]) {
     // #1737: bail before doing any work for this page if the run was aborted.
@@ -879,8 +855,10 @@ 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.
-  // #2552: local endpoints auto-cap — see resolveEmbedConcurrency.
-  const CONCURRENCY = resolveEmbedConcurrency(staleOpts?.paceMaxConcurrency);
+  const BASE_CONCURRENCY = parseInt(process.env.GBRAIN_EMBED_CONCURRENCY || '20', 10);
+  const CONCURRENCY = staleOpts?.paceMaxConcurrency
+    ? Math.min(BASE_CONCURRENCY, staleOpts.paceMaxConcurrency)
+    : BASE_CONCURRENCY;
   const pacer = staleOpts?.pacer ?? createNoopPacer();
 
   // D3 + D3a + D8: wall-clock budget. 30 min default; env override.

src/core/ai/gateway.ts

@@ -683,33 +683,6 @@ 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
@@ -1027,9 +1000,24 @@ const voyageCompatFetch = (async (input: RequestInfo | URL, init?: RequestInit)
   // Voyage diverges from OpenAI in two places that break the parser:
   //   - `embedding` is a base64 string (SDK schema expects `number[]`)
   //   - `usage` lacks `prompt_tokens` (SDK schema requires it when usage present)
+  //
+  // #1610: read the body ONCE via text() and JSON.parse it. The pre-fix
+  // `await resp.clone().json()` truncated large bodies on bun < 1.1.27
+  // (oven-sh/bun#6348) — the parse threw, the catch fell back to the raw
+  // response, and multi-chunk pages died with "Invalid JSON response".
+  // Every JSON return path below rebuilds the Response so a stale
+  // Content-Length/Content-Encoding header from the original can't lie
+  // about the rewritten body.
+  const bodyText = await resp.text();
+  const rebuild = (body: string) => {
+    const headers = new Headers(resp.headers);
+    headers.delete('content-length');
+    headers.delete('content-encoding');
+    return new Response(body, { status: resp.status, statusText: resp.statusText, headers });
+  };
   try {
-    const json: any = await resp.clone().json();
-    if (!json || typeof json !== 'object') return resp;
+    const json: any = JSON.parse(bodyText);
+    if (!json || typeof json !== 'object') return rebuild(bodyText);
     let modified = false;
     if (Array.isArray(json.data)) {
       for (const item of json.data) {
@@ -1064,22 +1052,19 @@ const voyageCompatFetch = (async (input: RequestInfo | URL, init?: RequestInit)
         : 0;
       modified = true;
     }
-    if (!modified) return resp;
-    return new Response(JSON.stringify(json), {
-      status: resp.status,
-      statusText: resp.statusText,
-      headers: resp.headers,
-    });
+    if (!modified) return rebuild(bodyText);
+    return rebuild(JSON.stringify(json));
   } catch (err) {
     // OOM-cap throws MUST propagate. The catch is here for "Voyage returned
     // JSON I can't reshape" (parse error, unexpected schema) — falling back
-    // to the original response is correct in that case. Letting the
+    // to the original body is correct in that case. Letting the
     // too-large response through here would defeat the entire purpose of
     // Layer 2 (the per-embedding cap that fires when Content-Length wasn't
     // available to Layer 1).
     if (err instanceof VoyageResponseTooLargeError) throw err;
-    // If parsing/transformation fails, fall back to the original response.
-    return resp;
+    // If parsing/transformation fails, pass the original body through
+    // (rebuilt — resp's body stream is already consumed by text()).
+    return rebuild(bodyText);
   }
 }) as unknown as typeof fetch;
 
@@ -1219,9 +1204,21 @@ const zeroEntropyCompatFetch = (async (input: RequestInfo | URL, init?: RequestI
   // validates. Also map usage.total_tokens → prompt_tokens (SDK requires
   // prompt_tokens when `usage` is present — same divergence Voyage hit at
   // gateway.ts:655).
+  //
+  // #1610: read the body ONCE via text() + JSON.parse — `resp.clone().json()`
+  // truncated large bodies on bun < 1.1.27 (oven-sh/bun#6348), so the parse
+  // threw and the catch fell back to the RAW ZE `{results: ...}` shape, which
+  // the AI SDK schema rejects → "Invalid JSON response" on multi-chunk pages.
+  const bodyText = await resp.text();
+  const rebuild = (body: string) => {
+    const headers = new Headers(resp.headers);
+    headers.delete('content-length');
+    headers.delete('content-encoding');
+    return new Response(body, { status: resp.status, statusText: resp.statusText, headers });
+  };
   try {
-    const json: any = await resp.clone().json();
-    if (!json || typeof json !== 'object') return resp;
+    const json: any = JSON.parse(bodyText);
+    if (!json || typeof json !== 'object') return rebuild(bodyText);
     let modified = false;
     if (Array.isArray(json.results) && !Array.isArray(json.data)) {
       // Layer 2 OOM cap — per-embedding size. ZE returns float[] arrays,
@@ -1255,20 +1252,25 @@ const zeroEntropyCompatFetch = (async (input: RequestInfo | URL, init?: RequestI
       // SDK also expects total_tokens; ZE provides it directly.
       modified = true;
     }
-    if (!modified) return resp;
-    return new Response(JSON.stringify(json), {
-      status: resp.status,
-      statusText: resp.statusText,
-      headers: resp.headers,
-    });
+    if (!modified) return rebuild(bodyText);
+    return rebuild(JSON.stringify(json));
   } catch (err) {
     // OOM-cap throws MUST propagate. Voyage's pattern: instanceof check on
     // its own tagged class. Same here — only rethrow our own cap class.
     if (err instanceof ZeroEntropyResponseTooLargeError) throw err;
-    return resp;
+    return rebuild(bodyText);
   }
 }) as unknown as typeof fetch;
 
+/**
+ * Test-only seams (#1610): the compat shims are module-private closures;
+ * exporting them lets tests drive the response-rewrite paths behaviorally
+ * (truncating clone(), stale Content-Length) without a live provider.
+ * Same pattern as __getShrinkStateForTests.
+ */
+export const __voyageCompatFetchForTests = voyageCompatFetch;
+export const __zeroEntropyCompatFetchForTests = zeroEntropyCompatFetch;
+
 /**
  * Generic asymmetric-embedding shim for openai-compatible recipes that
  * ship no compat fetch of their own (llama-server, litellm, ollama, ...).

src/core/ai/recipes/ollama.ts

@@ -29,17 +29,9 @@ 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',
-      // #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,
+      // 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,
     },
   },
   setup_hint: 'Install Ollama from https://ollama.ai, then `ollama pull nomic-embed-text` and `ollama serve`.',

src/core/doctor-categories.ts

@@ -141,7 +141,6 @@ 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/operations.ts

@@ -499,7 +499,7 @@ export function linkReadScopeOpts(ctx: OperationContext): { sourceId?: string; s
  * FAIL-CLOSED: anything not strictly `ctx.remote === false` is untrusted.
  *
  * This is the SINGLE resolver for every read op that accepts a per-call
- * `source_id` / `all_sources` parameter (query, code_callers, code_callees,
+ * `source_id` / `all_sources` parameter (query, search, code_callers, code_callees,
  * get_page, search_by_image, code_blast, code_flow). Inlining the `__all__`
  * branch per handler is the bug class that leaked cross-source reads (#1924,
  * #1371): a remote client could pass `source_id: '__all__'` to opt out of its
@@ -1442,13 +1442,24 @@ const search: Operation = {
     limit: { type: 'number', description: 'Max results (default 20)' },
     offset: { type: 'number', description: 'Skip first N results (for pagination)' },
     mode: { type: 'string', description: 'Search mode (conservative|balanced|tokenmax). Local callers only.' },
+    source_id: {
+      type: 'string',
+      description:
+        "Scope search to a single source. Defaults to OperationContext.sourceId. Pass '__all__' to span every source for trusted local callers; for remote callers '__all__' spans only your granted sources.",
+    },
+    all_sources: { type: 'boolean', description: "Span sources (equivalent to source_id=__all__): every source locally, your grant remotely." },
   },
   handler: async (ctx, p) => {
     const startedAt = Date.now();
     const queryText = p.query as string;
     const limit = (p.limit as number) || 20;
     const offset = (p.offset as number) || 0;
-    const scope = sourceScopeOpts(ctx);
+    // #1484 follow-up: route through the canonical fail-closed resolver so
+    // `--source-id __all__` / `all_sources` behave the same as on `query`
+    // (the zero-hit CLI hint advises exactly that retry). Without a per-call
+    // param, `search` silently ignored --source-id — the retry looked like
+    // a genuine miss.
+    const scope = resolveRequestedScope(ctx, p.source_id as string | undefined, p.all_sources === true);
 
     // T4/D5 — per-call mode honored ONLY for trusted/local callers so a remote
     // OAuth client can't escalate to the costly tokenmax bundle. Local + 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, buildWebsearchQueryExpr } from './search/sql-ranking.ts';
+import { buildSourceFactorCase, buildHardExcludeClause, buildVisibilityClause, buildRecencyComponentSql, buildBestPerPagePoolCte, buildOrFallbackWebsearchQuery } from './search/sql-ranking.ts';
 import {
   normalizeEngineColumn,
   buildVectorCastFragment,
@@ -1591,8 +1591,6 @@ 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) {
@@ -1632,7 +1630,6 @@ 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 (
@@ -1640,14 +1637,14 @@ 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, ${ftsQueryExpr}) * ${sourceFactorCase} AS score,
+           ts_rank(cc.search_vector, websearch_to_tsquery('${ftsLang}', $1)) * ${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 @@ ${ftsQueryExpr} ${detailFilter}${extraFilter} ${hardExcludeClause} ${visibilityClause}
+         WHERE cc.search_vector @@ websearch_to_tsquery('${ftsLang}', $1) ${detailFilter}${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.
@@ -1715,10 +1712,7 @@ 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);
 
-    // #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) {
@@ -1766,7 +1760,7 @@ 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, ${ftsQueryExpr}) * ${sourceFactorCase} AS score,
+         ts_rank_cd(p.search_vector, websearch_to_tsquery('${ftsLang}', $1)) * ${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
@@ -1781,7 +1775,7 @@ 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 @@ ${ftsQueryExpr}
+       WHERE p.search_vector @@ websearch_to_tsquery('${ftsLang}', $1)
          ${extraFilter} ${hardExcludeClause} ${visibilityClause}
        ORDER BY score DESC, p.id ASC
        LIMIT $2 OFFSET $3`;
@@ -1968,8 +1962,6 @@ 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) {
@@ -2004,21 +1996,20 @@ 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, ${ftsQueryExpr}) * ${sourceFactorCase} AS score,
+         ts_rank(cc.search_vector, websearch_to_tsquery('${ftsLang}', $1)) * ${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 @@ ${ftsQueryExpr} ${detailFilter}${extraFilter} ${hardExcludeClause} ${visibilityClause}
+       WHERE cc.search_vector @@ websearch_to_tsquery('${ftsLang}', $1) ${detailFilter}${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, buildWebsearchQueryExpr } from './search/sql-ranking.ts';
+import { buildSourceFactorCase, buildHardExcludeClause, buildVisibilityClause, buildRecencyComponentSql, buildBestPerPagePoolCte, buildOrFallbackWebsearchQuery } from './search/sql-ranking.ts';
 import { DEFAULT_EMBEDDING_MODEL, DEFAULT_EMBEDDING_DIMENSIONS } from './ai/defaults.ts';
 import { DELETE_BATCH_SIZE } from './engine-constants.ts';
 
@@ -1691,8 +1691,6 @@ 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) {
@@ -1763,7 +1761,6 @@ 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 (
@@ -1771,11 +1768,11 @@ 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, ${ftsQueryExpr}) * ${sourceFactorCase} AS score
+          ts_rank(cc.search_vector, websearch_to_tsquery('${ftsLang}', $1)) * ${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 @@ ${ftsQueryExpr}
+        WHERE cc.search_vector @@ websearch_to_tsquery('${ftsLang}', $1)
           ${typeClause}
           ${typesClause}
           ${excludeSlugsClause}
@@ -1866,10 +1863,7 @@ 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);
 
-    // #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) {
@@ -1929,7 +1923,7 @@ 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, ${ftsQueryExpr}) * ${sourceFactorCase} AS score,
+        ts_rank_cd(p.search_vector, websearch_to_tsquery('${ftsLang}', $1)) * ${sourceFactorCase} AS score,
         false AS stale
       FROM pages p
       JOIN sources s ON s.id = p.source_id
@@ -1942,7 +1936,7 @@ 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 @@ ${ftsQueryExpr}
+      WHERE p.search_vector @@ websearch_to_tsquery('${ftsLang}', $1)
         ${typeClause}
         ${typesClause}
         ${excludeSlugsClause}
@@ -2006,8 +2000,6 @@ 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) {
@@ -2068,19 +2060,18 @@ 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, ${ftsQueryExpr}) * ${sourceFactorCase} AS score,
+        ts_rank(cc.search_vector, websearch_to_tsquery('${ftsLang}', $1)) * ${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 @@ ${ftsQueryExpr}
+      WHERE cc.search_vector @@ websearch_to_tsquery('${ftsLang}', $1)
         ${typeClause}
         ${typesClause}
         ${excludeSlugsClause}

src/core/search/hybrid.ts

@@ -1323,8 +1323,18 @@ export async function hybridSearch(
       if (effectiveModality === 'both' && imageVectorList !== null) {
         vectorLists = [...vectorLists, imageVectorList];
       }
-    } catch {
-      // Embedding failure is non-fatal, fall back to keyword-only
+    } catch (err) {
+      // Embedding/vector failure is non-fatal — fall back to keyword-only —
+      // but say WHY (#1626): this arm only runs when the embedding provider
+      // probed available, so a throw here is a real failure (embed timeout,
+      // transient pooler error on the searchVector fan-out). Pre-fix the bare
+      // catch made a cross-source `--source __all__` run silently collapse to
+      // keyword-only/"No results" with zero diagnostics.
+      warnOncePerProcess(
+        'hybrid-vector-arm-failed',
+        `[gbrain] vector arm failed (fail-open, keyword-only fallback): ` +
+          `${err instanceof Error ? err.message : String(err)}`,
+      );
     }
   }
 

src/core/search/sql-ranking.ts

@@ -251,28 +251,6 @@ 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/compat-fetch-body-read.serial.test.ts

@@ -0,0 +1,115 @@
+/**
+ * #1610 — Voyage/ZeroEntropy compat shims must read the response body ONCE
+ * via text() instead of `resp.clone().json()`.
+ *
+ * On bun < 1.1.27, Response.clone() truncates large bodies (oven-sh/bun#6348):
+ * the clone().json() parse threw, the shim's catch fell back to the ORIGINAL
+ * response — whose wire shape (ZE `{results: ...}`, Voyage base64 embeddings)
+ * the AI SDK's openai-compatible Zod schema rejects — and multi-chunk pages
+ * failed with "Invalid JSON response".
+ *
+ * These tests simulate the truncating clone() and assert the shims still
+ * return the fully rewritten body. They also pin that the rewritten Response
+ * does NOT carry the original (now stale) Content-Length header, which lied
+ * about the rewritten body's size (gateway.ts previously copied
+ * `headers: resp.headers` verbatim).
+ */
+
+import { afterEach, describe, expect, test } from 'bun:test';
+import {
+  __voyageCompatFetchForTests,
+  __zeroEntropyCompatFetchForTests,
+} from '../../src/core/ai/gateway.ts';
+
+const origFetch = globalThis.fetch;
+afterEach(() => {
+  globalThis.fetch = origFetch;
+});
+
+/** Build a Response whose clone() truncates the body (bun < 1.1.27 behavior). */
+function truncatingCloneResponse(body: string): Response {
+  const headers = {
+    'content-type': 'application/json',
+    // Deliberately stale after any rewrite: the original wire body's length.
+    'content-length': String(Buffer.byteLength(body)),
+  };
+  const resp = new Response(body, { status: 200, headers });
+  (resp as any).clone = () =>
+    new Response(body.slice(0, 32), { status: 200, headers });
+  return resp;
+}
+
+describe('voyageCompatFetch — single body read (#1610)', () => {
+  test('rewrites base64 embeddings even when clone() truncates the body', async () => {
+    const floats = new Float32Array([0.5, 0.25, -1]);
+    const b64 = Buffer.from(floats.buffer).toString('base64');
+    const wireBody = JSON.stringify({
+      object: 'list',
+      data: [{ object: 'embedding', embedding: b64, index: 0 }],
+      model: 'voyage-3',
+      usage: { total_tokens: 7 },
+    });
+    globalThis.fetch = (async () => truncatingCloneResponse(wireBody)) as unknown as typeof fetch;
+
+    const out = await __voyageCompatFetchForTests('https://api.voyageai.com/v1/embeddings', {
+      method: 'POST',
+      body: JSON.stringify({ input: ['hello'], model: 'voyage-3' }),
+      headers: { 'content-type': 'application/json' },
+    });
+
+    const json: any = await out.json();
+    expect(Array.from(json.data[0].embedding)).toEqual([0.5, 0.25, -1]);
+    expect(json.usage.prompt_tokens).toBe(7);
+    // Stale Content-Length from the wire body must not survive the rewrite.
+    expect(out.headers.get('content-length')).toBeNull();
+    expect(out.headers.get('content-encoding')).toBeNull();
+  });
+});
+
+describe('zeroEntropyCompatFetch — single body read (#1610)', () => {
+  test('rewrites {results} → {data} even when clone() truncates the body', async () => {
+    const wireBody = JSON.stringify({
+      results: [{ embedding: [0.1, 0.2] }, { embedding: [0.3, 0.4] }],
+      usage: { total_bytes: 42, total_tokens: 9 },
+    });
+    let fetchedUrl = '';
+    globalThis.fetch = (async (url: string | URL | Request) => {
+      fetchedUrl = String(url);
+      return truncatingCloneResponse(wireBody);
+    }) as unknown as typeof fetch;
+
+    const out = await __zeroEntropyCompatFetchForTests('https://api.zeroentropy.dev/v1/embeddings', {
+      method: 'POST',
+      body: JSON.stringify({ input: ['hello'], model: 'zembed-1' }),
+      headers: { 'content-type': 'application/json' },
+    });
+
+    expect(fetchedUrl.endsWith('/v1/models/embed')).toBe(true);
+    const json: any = await out.json();
+    // The AI SDK schema requires {data: [{embedding, index}]} — the raw ZE
+    // {results} fallback is exactly the pre-fix "Invalid JSON response".
+    expect(json.results).toBeUndefined();
+    expect(json.data).toHaveLength(2);
+    expect(json.data[0]).toEqual({ object: 'embedding', embedding: [0.1, 0.2], index: 0 });
+    expect(json.data[1].index).toBe(1);
+    expect(json.usage.prompt_tokens).toBe(9);
+    expect(out.headers.get('content-length')).toBeNull();
+  });
+
+  test('non-JSON body falls back to the original bytes (rebuilt, still readable)', async () => {
+    const wireBody = 'plain text, not json';
+    globalThis.fetch = (async () =>
+      new Response(wireBody, {
+        status: 200,
+        headers: { 'content-type': 'application/json' },
+      })) as unknown as typeof fetch;
+
+    const out = await __zeroEntropyCompatFetchForTests('https://api.zeroentropy.dev/v1/embeddings', {
+      method: 'POST',
+      body: JSON.stringify({ input: ['hello'] }),
+    });
+    // Body was consumed by the shim's single read; the fallback must
+    // rebuild a readable Response rather than return the drained original.
+    expect(await out.text()).toBe(wireBody);
+  });
+});

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('LiteLLM and llama-server declare no_batch_cap: true', () => {
-    for (const id of ['litellm', 'llama-server']) {
+  test('Ollama, LiteLLM, llama-server all declare no_batch_cap: true', () => {
+    for (const id of ['ollama', 'litellm', 'llama-server']) {
       const r = getRecipe(id);
       expect(r, `${id} not registered`).toBeDefined();
       expect(
@@ -39,18 +39,6 @@ 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/ai/zeroentropy-compat-fetch.test.ts

@@ -98,16 +98,18 @@ describe('zeroEntropyCompatFetch — OOM caps', () => {
     expect(src).toMatch(/MAX_ZEROENTROPY_RESPONSE_BYTES\s*=\s*256\s*\*\s*1024\s*\*\s*1024/);
   });
 
-  test('Layer 1: Content-Length pre-check before resp.clone().json()', async () => {
+  test('Layer 1: Content-Length pre-check before the body is read', async () => {
     const src = await Bun.file(GATEWAY_PATH).text();
     // Find the zeroEntropyCompatFetch block bounds, then assert ordering
-    // within it (mirroring the voyage cap test pattern).
+    // within it (mirroring the voyage cap test pattern). #1610 moved the
+    // body read from `resp.clone().json()` to a single `resp.text()` (bun
+    // < 1.1.27 truncates clone()d bodies, oven-sh/bun#6348).
     const zeFetchStart = src.indexOf('const zeroEntropyCompatFetch');
     expect(zeFetchStart).toBeGreaterThan(0);
-    const block = src.slice(zeFetchStart, zeFetchStart + 8000);
+    const block = src.slice(zeFetchStart, zeFetchStart + 9000);
 
     const preCheckIdx = block.indexOf("resp.headers.get('content-length')");
-    const jsonParseIdx = block.indexOf('await resp.clone().json()');
+    const jsonParseIdx = block.indexOf('const bodyText = await resp.text()');
     expect(preCheckIdx).toBeGreaterThan(0);
     expect(jsonParseIdx).toBeGreaterThan(0);
     // The pre-check MUST appear before the JSON parse — Voyage's lesson

test/cli-source-scope-hint.test.ts

@@ -0,0 +1,61 @@
+/**
+ * #1484 — invisible-miss hint. A bare `gbrain query` resolves to a single
+ * source (usually 'default'); on a multi-source brain a zero-hit run gave no
+ * signal that the answer might live in another source. sourceScopeHint
+ * returns the stderr hint exactly when: query/search op + zero results +
+ * no explicit scoping param + >1 registered source.
+ */
+
+import { describe, expect, test } from 'bun:test';
+import { sourceScopeHint } from '../src/cli.ts';
+import type { BrainEngine } from '../src/core/engine.ts';
+
+function fakeEngine(sourceCount: number, fail = false): BrainEngine {
+  return {
+    executeRaw: async () => {
+      if (fail) throw new Error('sources table missing');
+      return [{ n: sourceCount }];
+    },
+  } as unknown as BrainEngine;
+}
+
+describe('sourceScopeHint (#1484)', () => {
+  test('fires on a bare zero-hit query against a multi-source brain', async () => {
+    const hint = await sourceScopeHint('query', {}, 'default', fakeEngine(3), []);
+    expect(hint).toContain('3 sources');
+    expect(hint).toContain('"default"');
+    expect(hint).toContain('--source-id __all__');
+  });
+
+  test('fires for search too', async () => {
+    const hint = await sourceScopeHint('search', {}, 'wiki', fakeEngine(2), []);
+    expect(hint).toContain('"wiki"');
+  });
+
+  test('silent when results were found', async () => {
+    expect(await sourceScopeHint('query', {}, 'default', fakeEngine(3), [{ slug: 'a' }])).toBeNull();
+  });
+
+  test('silent when the caller scoped explicitly', async () => {
+    expect(await sourceScopeHint('query', { source_id: 'wiki' }, 'wiki', fakeEngine(3), [])).toBeNull();
+    expect(await sourceScopeHint('query', { source: 'wiki' }, 'wiki', fakeEngine(3), [])).toBeNull();
+    expect(await sourceScopeHint('query', { all_sources: true }, '__all__', fakeEngine(3), [])).toBeNull();
+  });
+
+  test('silent when the resolved scope is already __all__', async () => {
+    expect(await sourceScopeHint('query', {}, '__all__', fakeEngine(3), [])).toBeNull();
+  });
+
+  test('silent on a single-source brain', async () => {
+    expect(await sourceScopeHint('query', {}, 'default', fakeEngine(1), [])).toBeNull();
+  });
+
+  test('silent for non-search ops and non-array results', async () => {
+    expect(await sourceScopeHint('get_stats', {}, 'default', fakeEngine(3), [])).toBeNull();
+    expect(await sourceScopeHint('query', {}, 'default', fakeEngine(3), { rows: [] })).toBeNull();
+  });
+
+  test('best-effort: sources lookup failure returns null, never throws', async () => {
+    expect(await sourceScopeHint('query', {}, 'default', fakeEngine(3, true), [])).toBeNull();
+  });
+});

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

@@ -1,118 +0,0 @@
-/**
- * #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/hybrid-vector-arm-warn.serial.test.ts

@@ -0,0 +1,77 @@
+/**
+ * #1626 — hybridSearch's text-vector arm must not fail DARK.
+ *
+ * The arm only runs when the embedding provider probed available, so a throw
+ * inside it (embed timeout, transient pooler error on searchVector) is a real
+ * failure. Pre-fix, a bare `catch {}` swallowed it and the run silently
+ * collapsed to keyword-only — under `--source __all__` on a strained pooler
+ * that read as a non-deterministic "No results". The fix logs the swallowed
+ * reason via warnOncePerProcess while keeping the keyword fallback.
+ */
+
+import { afterAll, beforeAll, describe, expect, test } from 'bun:test';
+import { PGLiteEngine } from '../src/core/pglite-engine.ts';
+import { hybridSearch } from '../src/core/search/hybrid.ts';
+import {
+  __setEmbedTransportForTests,
+  configureGateway,
+  resetGateway,
+} from '../src/core/ai/gateway.ts';
+import { _resetWarnOnceForTests } from '../src/core/utils.ts';
+
+let engine: PGLiteEngine;
+const origWarn = console.warn;
+
+beforeAll(async () => {
+  // Pin the gateway to OpenAI with a stub key (put-page-provenance pattern):
+  // embed() runs instantiateEmbedding — which requires OPENAI_API_KEY — BEFORE
+  // the stubbed transport is reached. Without this, a keyless CI environment
+  // throws the config error instead of the transport's, and the assertion on
+  // the swallowed reason fails. The key never leaves the process.
+  configureGateway({
+    embedding_model: 'openai:text-embedding-3-large',
+    embedding_dimensions: 1536,
+    env: { ...process.env, OPENAI_API_KEY: process.env.OPENAI_API_KEY || 'sk-test-stub' },
+  });
+  engine = new PGLiteEngine();
+  await engine.connect({});
+  await engine.initSchema();
+  await engine.putPage('people/alice-example', {
+    type: 'person',
+    title: 'Alice Example',
+    compiled_truth: 'Alice Example is a test person for the vector-arm warn test.',
+  });
+});
+
+afterAll(async () => {
+  console.warn = origWarn;
+  __setEmbedTransportForTests(null);
+  resetGateway();
+  await engine.disconnect();
+});
+
+describe('hybridSearch vector-arm failure telemetry (#1626)', () => {
+  test('embed failure logs the swallowed reason and falls back to keyword', async () => {
+    _resetWarnOnceForTests();
+    // Installing a transport makes isAvailable('embedding') true (test-seam
+    // fast path), so the vector arm RUNS — and then throws.
+    __setEmbedTransportForTests(() => {
+      throw new Error('pooler exploded mid-fanout');
+    });
+    const warnings: string[] = [];
+    console.warn = (...args: unknown[]) => {
+      warnings.push(args.map(String).join(' '));
+    };
+    try {
+      const results = await hybridSearch(engine, 'alice');
+      // Keyword fallback still returns results — fail-open preserved.
+      expect(results.some((r) => r.slug === 'people/alice-example')).toBe(true);
+    } finally {
+      console.warn = origWarn;
+      __setEmbedTransportForTests(null);
+    }
+    const armWarnings = warnings.filter((w) => w.includes('vector arm failed'));
+    expect(armWarnings).toHaveLength(1);
+    expect(armWarnings[0]).toContain('pooler exploded mid-fanout');
+  });
+});

test/pglite-engine.test.ts

@@ -216,67 +216,6 @@ 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

test/search-op-source-scope.test.ts

@@ -0,0 +1,87 @@
+/**
+ * #1484 follow-up — the `search` op must honor per-call `source_id` /
+ * `all_sources` through the canonical fail-closed resolver
+ * (resolveRequestedScope), exactly like `query` does.
+ *
+ * Pre-fix, `search` had no source_id param at all: the zero-hit CLI hint
+ * advised "retry with --source-id __all__", the flag parsed into params,
+ * NOTHING consumed it, and the retry silently re-ran the same single-source
+ * search — an invisible false negative (and the retry's params.source_id
+ * suppressed the hint, so the user got no second warning).
+ */
+
+import { describe, expect, test } from 'bun:test';
+import { operationsByName } from '../src/core/operations.ts';
+import type { OperationContext } from '../src/core/operations.ts';
+import type { BrainEngine } from '../src/core/engine.ts';
+
+const searchOp = operationsByName['search'];
+
+/** Fake engine: keyword-only config so the handler's scope goes straight to
+ *  searchKeyword, where we capture the opts it was called with. */
+function makeCtx(remote: boolean, allowedSources?: string[]) {
+  const captured: { opts?: Record<string, unknown> } = {};
+  const engine = {
+    getConfig: async (key: string) => (key === 'search.mcp_keyword_only' ? 'true' : null),
+    searchKeyword: async (_q: string, opts: Record<string, unknown>) => {
+      captured.opts = opts;
+      return [];
+    },
+  } as unknown as BrainEngine;
+  const ctx = {
+    engine,
+    config: { engine: 'pglite' },
+    logger: { info: () => {}, warn: () => {}, error: () => {} },
+    dryRun: false,
+    remote,
+    sourceId: 'default',
+    ...(allowedSources ? { auth: { allowedSources } } : {}),
+  } as unknown as OperationContext;
+  return { ctx, captured };
+}
+
+describe('search op per-call source scope (#1484 follow-up)', () => {
+  test('op declares source_id + all_sources params (the CLI hint advises them)', () => {
+    expect(searchOp.params.source_id).toBeDefined();
+    expect(searchOp.params.all_sources).toBeDefined();
+  });
+
+  test('default: scopes to ctx.sourceId', async () => {
+    const { ctx, captured } = makeCtx(false);
+    await searchOp.handler(ctx, { query: 'x' });
+    expect(captured.opts?.sourceId).toBe('default');
+  });
+
+  test("local + source_id '__all__' spans the whole brain (no source filter)", async () => {
+    const { ctx, captured } = makeCtx(false);
+    await searchOp.handler(ctx, { query: 'x', source_id: '__all__' });
+    expect(captured.opts?.sourceId).toBeUndefined();
+    expect(captured.opts?.sourceIds).toBeUndefined();
+  });
+
+  test('local + all_sources=true spans the whole brain', async () => {
+    const { ctx, captured } = makeCtx(false);
+    await searchOp.handler(ctx, { query: 'x', all_sources: true });
+    expect(captured.opts?.sourceId).toBeUndefined();
+    expect(captured.opts?.sourceIds).toBeUndefined();
+  });
+
+  test('explicit source_id wins over ctx.sourceId', async () => {
+    const { ctx, captured } = makeCtx(false);
+    await searchOp.handler(ctx, { query: 'x', source_id: 'wiki' });
+    expect(captured.opts?.sourceId).toBe('wiki');
+  });
+
+  test("remote + '__all__' collapses to the caller's grant (fail-closed)", async () => {
+    const { ctx, captured } = makeCtx(true, ['wiki', 'essays']);
+    await searchOp.handler(ctx, { query: 'x', source_id: '__all__' });
+    expect(captured.opts?.sourceIds).toEqual(['wiki', 'essays']);
+  });
+
+  test('remote + out-of-grant source_id is denied', async () => {
+    const { ctx } = makeCtx(true, ['wiki']);
+    await expect(searchOp.handler(ctx, { query: 'x', source_id: 'secrets' })).rejects.toThrow(
+      /outside your granted sources/,
+    );
+  });
+});

test/voyage-response-cap.test.ts

@@ -34,7 +34,7 @@ describe('v0.31.8 — voyage Content-Length pre-check + per-item cap', () => {
     expect(source).toMatch(/MAX_VOYAGE_RESPONSE_BYTES\s*=\s*256\s*\*\s*1024\s*\*\s*1024/);
   });
 
-  test('Layer 1: Content-Length pre-check fires BEFORE resp.clone().json() (D10 OOM defense)', async () => {
+  test('Layer 1: Content-Length pre-check fires BEFORE the body is read (D10 OOM defense)', async () => {
     const source = await Bun.file(new URL('../src/core/ai/gateway.ts', import.meta.url)).text();
     // Anchor relative to the post-fetch handler block. The function declaration
     // contains an OUTBOUND request body section earlier; we want to verify
@@ -47,8 +47,10 @@ describe('v0.31.8 — voyage Content-Length pre-check + per-item cap', () => {
     // doesn't pin to comment text.
     const preCheckIdx = inboundBlock.indexOf("resp.headers.get('content-length')");
     // Use the full lvalue assignment so the match doesn't accidentally hit
-    // comment text that mentions `await resp.clone().json()` for context.
-    const jsonParseIdx = inboundBlock.indexOf('const json: any = await resp.clone().json()');
+    // comment text that mentions the body read for context. (#1610 moved the
+    // read from `resp.clone().json()` to a single `resp.text()` — bun <
+    // 1.1.27 truncates clone()d bodies, oven-sh/bun#6348.)
+    const jsonParseIdx = inboundBlock.indexOf('const bodyText = await resp.text()');
     expect(preCheckIdx).toBeGreaterThan(0);
     expect(jsonParseIdx).toBeGreaterThan(0);
     // The pre-check MUST appear before the JSON parse — otherwise the OOM