feat(ai): add reranker touchpoint to LiteLLM proxy recipe (takeover of #2455)

LiteLLM normalizes Cohere/Voyage/Jina rerank backends to the wire shape
gateway.rerank() already speaks, so a reranker touchpoint on the litellm
recipe makes any proxied rerank model reachable via
`search.reranker.model litellm:<model>` with no adapter.

Repairs from the original PR:
- path is the LEAF '/rerank' (not '/v1/rerank'): LiteLLM serves both
  /rerank and /v1/rerank, and the recipe's setup_hint allows
  LITELLM_BASE_URL with or without the /v1 suffix — pinning '/v1/rerank'
  doubled to /v1/v1/rerank (404) on /v1-suffixed bases.
- setup_hint appends the rerank guidance to master's current line instead
  of replacing it with a stale pre-/v1-suffix version.
- cost_per_1m_tokens_usd stays undefined (pricing-unknown), matching the
  recipe's embedding/chat touchpoints and budget-tracker's deliberate
  litellm exclusion from the free-provider sets (a proxy can front a paid
  provider; the touchpoint field isn't consumed by rerank pricing anyway).

Test drives gateway.rerank()'s real URL builder via the stubbed transport
for both base-URL forms; the /v1-suffixed case fails with the original
PR's path.

Co-authored-by: ozp <ozp@users.noreply.github.com>
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>

src/commands/migrate-engine.ts

@@ -3,11 +3,7 @@
  *
  * Usage:
  *   gbrain migrate --to supabase [--url <connection_string>]
- *     (--url is persisted to config.json, mode 0600, so the migrated brain
- *      works without env — #1271)
  *   gbrain migrate --to pglite [--path <db_path>]
- *     (an explicit --path destination is bootstrapped with its own
- *      <path>/.gbrain/config.json so GBRAIN_HOME=<path> just works — #1271)
  *   gbrain migrate --to <engine> --force  (overwrite non-empty target)
  */
 
@@ -15,9 +11,9 @@ import { createEngine } from '../core/engine-factory.ts';
 import { loadConfig, saveConfig, toEngineConfig, gbrainPath, effectiveEnvDatabaseUrl, type GBrainConfig } from '../core/config.ts';
 import type { BrainEngine } from '../core/engine.ts';
 import type { EngineConfig } from '../core/types.ts';
-import { writeFileSync, readFileSync, existsSync, unlinkSync, mkdirSync, chmodSync } from 'fs';
+import { writeFileSync, readFileSync, existsSync, unlinkSync } from 'fs';
 import { createHash } from 'crypto';
-import { resolve, join } from 'path';
+import { resolve } from 'path';
 import { createProgress } from '../core/progress.ts';
 import { getCliOptions, cliOptsToProgressOptions } from '../core/cli-options.ts';
 
@@ -63,31 +59,6 @@ export interface MigrateManifest {
   started_at: string;
 }
 
-/**
- * #1271 Finding 1: make an explicit `--to pglite --path P` destination usable
- * as a standalone brain. Writes `P/.gbrain/config.json` (mode 0600, plus a
- * `*` .gitignore) so `GBRAIN_HOME=P` resolves without a manual `gbrain init`.
- * Never clobbers an existing config at the destination. Returns the written
- * config path, or null when skipped.
- */
-export function bootstrapDestinationConfig(dbPath: string): string | null {
-  const abs = resolve(dbPath);
-  const dir = join(abs, '.gbrain');
-  const file = join(dir, 'config.json');
-  if (existsSync(file)) return null;
-  mkdirSync(dir, { recursive: true });
-  const cfg: GBrainConfig = { engine: 'pglite', database_path: abs };
-  writeFileSync(file, JSON.stringify(cfg, null, 2) + '\n', { mode: 0o600 });
-  try { chmodSync(file, 0o600); } catch { /* platform-specific */ }
-  // Same worktree-safety pattern as saveConfig()'s ensureGitignore, scoped
-  // to the destination home. Don't clobber a user-customized .gitignore.
-  const gitignore = join(dir, '.gitignore');
-  if (!existsSync(gitignore)) {
-    writeFileSync(gitignore, '*\n', { mode: 0o600 });
-  }
-  return file;
-}
-
 export function migrationTargetId(config: EngineConfig): string {
   const locator = config.engine === 'postgres'
     ? config.database_url ?? ''
@@ -381,25 +352,6 @@ export async function runMigrateEngine(sourceEngine: BrainEngine, args: string[]
   };
   saveConfig(newConfig);
 
-  // #1271 Finding 2 (by design, but say it out loud): the connection string
-  // is persisted so the migrated brain works without env. Mode 0600.
-  if (opts.targetEngine === 'postgres' && opts.targetUrl) {
-    console.error('Note: the --url connection string (including credentials) is persisted to config.json (mode 0600).');
-  }
-
-  // #1271 Finding 1: an explicit --path destination doubles as a standalone
-  // GBRAIN_HOME. Best-effort — never fail a completed migration over it.
-  if (opts.targetEngine === 'pglite' && opts.targetPath) {
-    try {
-      const written = bootstrapDestinationConfig(opts.targetPath);
-      if (written) {
-        console.log(`Destination bootstrapped: ${written} (usable via GBRAIN_HOME=${resolve(opts.targetPath)})`);
-      }
-    } catch (e) {
-      console.warn(`  WARN could not bootstrap destination config: ${e instanceof Error ? e.message : String(e)}`);
-    }
-  }
-
   // Clean up
   clearManifest();
 

src/core/ai/recipes/litellm-proxy.ts

@@ -56,6 +56,36 @@ export const litellmProxy: Recipe = {
       cost_per_1m_output_usd: undefined,
       price_last_verified: '2026-06-14',
     },
+    // LiteLLM normalizes Cohere / Voyage / Jina / etc. rerank backends to the
+    // same wire shape gbrain's gateway.rerank() already speaks (the
+    // ZeroEntropy/llama.cpp contract):
+    //   { model, query, documents, top_n } → { results: [{ index, relevance_score }] }
+    // So any rerank model the user registers in their LiteLLM config is
+    // reachable via `gbrain config set search.reranker.model litellm:<model>`
+    // with no request/response adapter — same as embeddings ride the proxy.
+    reranker: {
+      models: [], // user-provided; whatever rerank models the proxy serves
+      // No canonical default — the proxy defines its own model ids. The user
+      // sets search.reranker.model explicitly (mirrors the embedding
+      // touchpoint's user_provided_models contract).
+      default_model: '',
+      // The proxied backend bills (Cohere/Voyage/…); pricing-unknown is the
+      // honest state — same stance as this recipe's embedding/chat
+      // touchpoints and budget-tracker's deliberate litellm exclusion from
+      // the free-provider sets.
+      cost_per_1m_tokens_usd: undefined,
+      price_last_verified: '2026-06-27',
+      max_payload_bytes: 5_000_000,
+      // LEAF path only (matches llama-server-reranker's convention). LiteLLM
+      // serves both `/rerank` and `/v1/rerank`, and LITELLM_BASE_URL may be
+      // set with or without the `/v1` suffix (the setup_hint allows both), so
+      // the leaf form yields a valid route either way:
+      //   http://localhost:4000    + /rerank → /rerank        ✓
+      //   http://localhost:4000/v1 + /rerank → /v1/rerank     ✓
+      // Pinning '/v1/rerank' here would double to /v1/v1/rerank → 404 on
+      // /v1-suffixed bases.
+      path: '/rerank',
+    },
   },
-  setup_hint: 'Run LiteLLM (https://docs.litellm.ai) in front of any provider; set LITELLM_BASE_URL (include the /v1 suffix if your proxy serves the OpenAI route there, e.g. http://localhost:4000/v1) + pass --embedding-model litellm:<model> and --embedding-dimensions <N>.',
+  setup_hint: 'Run LiteLLM (https://docs.litellm.ai) in front of any provider; set LITELLM_BASE_URL (include the /v1 suffix if your proxy serves the OpenAI route there, e.g. http://localhost:4000/v1) + pass --embedding-model litellm:<model> and --embedding-dimensions <N>. For rerank: register a rerank model in LiteLLM and set search.reranker.model litellm:<model-name>.',
 };

test/ai/recipe-litellm-reranker.test.ts

@@ -0,0 +1,88 @@
+/**
+ * litellm-proxy reranker touchpoint smoke.
+ *
+ * Sibling of recipe-llama-server-reranker.test.ts. Pins the reranker
+ * touchpoint on the LiteLLM proxy recipe so:
+ *  - the touchpoint exists with the LEAF '/rerank' path (LiteLLM serves both
+ *    /rerank and /v1/rerank, so the leaf form is valid whether or not the
+ *    user's LITELLM_BASE_URL carries the /v1 suffix the setup_hint allows)
+ *  - a /v1-suffixed base URL does NOT produce /v1/v1/rerank (the original
+ *    community PR pinned '/v1/rerank' which 404s on /v1-suffixed bases)
+ *  - models: [] (user-provided; proxy defines the model ids)
+ *  - pricing stays undefined (proxy can front a paid provider — same honest
+ *    pricing-unknown stance as the embedding/chat touchpoints)
+ *
+ * The gateway.rerank() URL tests drive the real URL builder via the stubbed
+ * transport (same seam as test/ai/rerank.test.ts).
+ */
+
+import { describe, expect, test, afterEach } from 'bun:test';
+import { getRecipe } from '../../src/core/ai/recipes/index.ts';
+import {
+  configureGateway,
+  resetGateway,
+  rerank,
+  __setRerankTransportForTests,
+} from '../../src/core/ai/gateway.ts';
+
+afterEach(() => {
+  __setRerankTransportForTests(null);
+  resetGateway();
+});
+
+describe('recipe: litellm reranker touchpoint', () => {
+  test('declares reranker touchpoint with leaf /rerank path', () => {
+    const r = getRecipe('litellm')!;
+    const tp = r.touchpoints.reranker;
+    expect(tp).toBeDefined();
+    expect(tp!.path).toBe('/rerank');
+    expect(tp!.max_payload_bytes).toBe(5_000_000);
+  });
+
+  test('reranker touchpoint uses empty models[] for user-provided model ids', () => {
+    const r = getRecipe('litellm')!;
+    expect(r.touchpoints.reranker!.models).toEqual([]);
+  });
+
+  test('pricing stays undefined — proxy can front a paid provider', () => {
+    const r = getRecipe('litellm')!;
+    expect(r.touchpoints.reranker!.cost_per_1m_tokens_usd).toBeUndefined();
+  });
+
+  test('setup_hint keeps the /v1-suffix guidance AND mentions rerank', () => {
+    const r = getRecipe('litellm')!;
+    expect(r.setup_hint).toMatch(/\/v1 suffix/);
+    expect(r.setup_hint).toMatch(/search\.reranker\.model litellm:/);
+  });
+});
+
+describe('gateway.rerank() URL via litellm recipe', () => {
+  async function capturedRerankUrl(baseUrl?: string): Promise<string> {
+    configureGateway({
+      reranker_model: 'litellm:my-reranker',
+      env: {},
+      ...(baseUrl ? { base_urls: { litellm: baseUrl } } : {}),
+    });
+    let capturedUrl = '';
+    __setRerankTransportForTests(async (url) => {
+      capturedUrl = url;
+      return new Response(
+        JSON.stringify({ results: [{ index: 0, relevance_score: 0.9 }] }),
+        { status: 200, headers: { 'content-type': 'application/json' } },
+      );
+    });
+    await rerank({ query: 'q', documents: ['d'] });
+    return capturedUrl;
+  }
+
+  test('default base (no /v1 suffix) → /rerank', async () => {
+    const url = await capturedRerankUrl();
+    expect(url).toBe('http://localhost:4000/rerank');
+  });
+
+  test('/v1-suffixed base → /v1/rerank, NOT /v1/v1/rerank', async () => {
+    const url = await capturedRerankUrl('http://localhost:4000/v1');
+    expect(url).toBe('http://localhost:4000/v1/rerank');
+    expect(url).not.toContain('/v1/v1/');
+  });
+});

test/migrate-engine-dest-bootstrap.test.ts

@@ -1,40 +0,0 @@
-import { describe, expect, test } from 'bun:test';
-import { mkdtempSync, mkdirSync, readFileSync, statSync, writeFileSync } from 'fs';
-import { tmpdir } from 'os';
-import { join, resolve } from 'path';
-import { bootstrapDestinationConfig } from '../src/commands/migrate-engine.ts';
-import { loadConfigFileOnly } from '../src/core/config.ts';
-import { withEnv } from './helpers/with-env.ts';
-
-describe('migrate --to pglite destination bootstrap (#1271)', () => {
-  test('writes <path>/.gbrain/config.json so GBRAIN_HOME=<path> resolves a brain', async () => {
-    const dest = mkdtempSync(join(tmpdir(), 'gbrain-dest-'));
-    const written = bootstrapDestinationConfig(dest);
-    const file = join(dest, '.gbrain', 'config.json');
-    expect(written).toBe(file);
-
-    const cfg = JSON.parse(readFileSync(file, 'utf-8'));
-    expect(cfg.engine).toBe('pglite');
-    expect(cfg.database_path).toBe(resolve(dest));
-    expect(statSync(file).mode & 0o777).toBe(0o600);
-    // worktree safety: destination home is git-ignored like saveConfig()'s home
-    expect(readFileSync(join(dest, '.gbrain', '.gitignore'), 'utf-8')).toBe('*\n');
-
-    // The exact failure mode from #1271: config resolution under
-    // GBRAIN_HOME=<path> used to find nothing ("No brain configured").
-    await withEnv({ GBRAIN_HOME: dest }, () => {
-      const loaded = loadConfigFileOnly();
-      expect(loaded?.engine).toBe('pglite');
-      expect(loaded?.database_path).toBe(resolve(dest));
-    });
-  });
-
-  test('never clobbers an existing destination config', () => {
-    const dest = mkdtempSync(join(tmpdir(), 'gbrain-dest-'));
-    mkdirSync(join(dest, '.gbrain'), { recursive: true });
-    writeFileSync(join(dest, '.gbrain', 'config.json'), '{"engine":"postgres"}\n');
-
-    expect(bootstrapDestinationConfig(dest)).toBe(null);
-    expect(JSON.parse(readFileSync(join(dest, '.gbrain', 'config.json'), 'utf-8')).engine).toBe('postgres');
-  });
-});