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/cli.ts

@@ -24,7 +24,6 @@ import type { GBrainConfig } from './core/config.ts';
 import type { AIGatewayConfig } from './core/ai/types.ts';
 import type { BrainEngine } from './core/engine.ts';
 import { operations, OperationError } from './core/operations.ts';
-import { resolveSourceIdEngineFree } from './core/source-resolver.ts';
 import { formatVolunteeredPage } from './core/context/volunteer.ts';
 import type { Operation, OperationContext } from './core/operations.ts';
 import { shouldForceExitAfterMain, finishCliTeardown, flushThenExit, currentExitCode, setCliExitVerdict } from './core/cli-force-exit.ts';
@@ -383,15 +382,6 @@ async function main() {
     if (op.localOnly) {
       refuseThinClient(command, cfgPre!.remote_mcp!.mcp_url);
     }
-    // #2098: the local path resolves --source / GBRAIN_SOURCE / .gbrain-source
-    // inside makeContext (ctx.sourceId), which this route never reaches — so
-    // scope must be mapped onto the op's source_id wire param before the call.
-    try {
-      applyThinClientSourceScope(op, params);
-    } catch (e: unknown) {
-      console.error(e instanceof Error ? e.message : String(e));
-      process.exit(1);
-    }
     await runThinClientRouted(op, params, cfgPre!, cliOpts);
     return;
   }
@@ -812,60 +802,6 @@ export function parseOpArgs(op: Operation, args: string[]): Record<string, unkno
   return params;
 }
 
-/**
- * #2098: thin-client source scoping. Locally, --source / GBRAIN_SOURCE /
- * .gbrain-source resolve to ctx.sourceId in makeContext; the thin-client
- * route short-circuits before that, so `gbrain query --source X` against a
- * remote brain silently searched unscoped. This runs the engine-free tiers
- * (flag → env → dotfile; the DB-backed tiers can't run without an engine —
- * the server's grant scoping covers the rest) and maps the result onto the
- * op's `source_id` wire param.
- *
- * Ops that declare their OWN `source` param (facts add, etc.) are left
- * untouched — their --source is an op param, not scope. An explicit --source
- * on an op with no source_id wire param throws (loud beats silent drop);
- * ambient env/dotfile scope with nowhere to send it is ignored, matching the
- * pre-fix behavior for non-scopeable ops. Exported for tests.
- */
-// Ops whose `source_id` wire param is NOT read-scope semantics: get_skill's
-// source_id flips the lookup from host catalog to brain-resident-pack
-// (getResidentSkillDetail). Ambient env/dotfile scope must never leak into
-// these; an explicit --source-id still passes through untouched above.
-const NON_SCOPE_SOURCE_ID_OPS = new Set(['get_skill']);
-
-export function applyThinClientSourceScope(
-  op: Operation,
-  params: Record<string, unknown>,
-  cwd?: string,
-): void {
-  if ('source' in op.params) return; // the op owns --source; not a scope flag
-  const explicit = typeof params.source === 'string' && params.source.length > 0
-    ? (params.source as string)
-    : null;
-  delete params.source; // never a wire param on these ops — don't leak it
-  // Explicit per-call scope already on the wire wins over ambient tiers.
-  if (params.source_id !== undefined || params.all_sources === true) {
-    if (explicit) {
-      throw new Error('Pass either --source or --source-id/--all-sources, not both.');
-    }
-    return;
-  }
-  const resolved = resolveSourceIdEngineFree(explicit, cwd);
-  if (!resolved) return;
-  if (!('source_id' in op.params) || NON_SCOPE_SOURCE_ID_OPS.has(op.name)) {
-    if (explicit) {
-      const hint = NON_SCOPE_SOURCE_ID_OPS.has(op.name)
-        ? `(its source_id parameter is not a scope filter; pass --source-id explicitly if you mean it)`
-        : `(the remote op has no source_id parameter; the server scopes it to your grant)`;
-      throw new Error(
-        `gbrain ${op.cliHints?.name || op.name} does not accept --source on a thin-client install ${hint}.`,
-      );
-    }
-    return; // ambient env/dotfile scope with nowhere to send it
-  }
-  params.source_id = resolved;
-}
-
 async function makeContext(engine: BrainEngine, params: Record<string, unknown>): Promise<OperationContext> {
   // v0.31.8 (D11): resolve sourceId via the canonical 6-tier chain. Honors
   // --source / GBRAIN_SOURCE / .gbrain-source / path-match / brain default /

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>.',
 };

src/core/source-resolver.ts

@@ -160,33 +160,6 @@ export async function resolveSourceId(
   return 'default';
 }
 
-/**
- * Engine-free tiers (1-3) of the resolution chain: explicit flag →
- * GBRAIN_SOURCE env → .gbrain-source dotfile walk. Used by the thin-client
- * CLI path (#2098), which has no local engine to run tiers 4-6 or
- * assertSourceExists against — the remote server enforces existence + grant.
- * Returns null when no engine-free tier fires.
- */
-export function resolveSourceIdEngineFree(
-  explicit: string | null | undefined,
-  cwd: string = process.cwd(),
-): string | null {
-  if (explicit) {
-    if (!SOURCE_ID_RE.test(explicit)) {
-      throw new Error(`Invalid --source value "${explicit}". Must match [a-z0-9-]{1,32}.`);
-    }
-    return explicit;
-  }
-  const env = process.env.GBRAIN_SOURCE;
-  if (env && env.length > 0) {
-    if (!SOURCE_ID_RE.test(env)) {
-      throw new Error(`Invalid GBRAIN_SOURCE value "${env}". Must match [a-z0-9-]{1,32}.`);
-    }
-    return env;
-  }
-  return readDotfileWalk(cwd);
-}
-
 /**
  * Returns the id of the SINGLE registered non-default source with a
  * local_path, when exactly one such row exists. Returns null when:

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/thin-client-source-scope.test.ts

@@ -1,142 +0,0 @@
-/**
- * #2098: thin-client routing dropped --source / GBRAIN_SOURCE / .gbrain-source.
- *
- * The local CLI path resolves source scope in makeContext (ctx.sourceId); the
- * thin-client route short-circuits before that and sent params verbatim, so
- * `gbrain query --source X` against a remote brain silently searched unscoped
- * (the server op ignores the unknown `source` key).
- *
- * applyThinClientSourceScope runs the engine-free tiers (flag → env → dotfile)
- * and maps the result onto the op's `source_id` wire param. These tests fail
- * without the fix (params.source_id stays undefined / params.source leaks).
- */
-
-import { describe, test, expect } from 'bun:test';
-import { mkdtempSync, rmSync, writeFileSync } from 'fs';
-import { join } from 'path';
-import { tmpdir } from 'os';
-import { applyThinClientSourceScope, parseOpArgs } from '../src/cli.ts';
-import { operationsByName } from '../src/core/operations.ts';
-import { withEnv } from './helpers/with-env.ts';
-
-const queryOp = operationsByName.query;
-
-describe('applyThinClientSourceScope (#2098)', () => {
-  test('--source maps onto the query op wire param source_id', async () => {
-    await withEnv({ GBRAIN_SOURCE: undefined }, () => {
-      const params = parseOpArgs(queryOp, ['find things', '--source', 'wiki']);
-      expect(params.source).toBe('wiki'); // pre-fix state: wrong key
-      applyThinClientSourceScope(queryOp, params, '/');
-      expect(params.source_id).toBe('wiki');
-      expect('source' in params).toBe(false); // never leaks the unknown key
-    });
-  });
-
-  test('GBRAIN_SOURCE env tier fires when no flag is passed', async () => {
-    await withEnv({ GBRAIN_SOURCE: 'gstack' }, () => {
-      const params = parseOpArgs(queryOp, ['find things']);
-      applyThinClientSourceScope(queryOp, params, '/');
-      expect(params.source_id).toBe('gstack');
-    });
-  });
-
-  test('.gbrain-source dotfile tier fires when flag and env are absent', async () => {
-    await withEnv({ GBRAIN_SOURCE: undefined }, () => {
-      const tmp = mkdtempSync(join(tmpdir(), 'gbrain-thin-scope-'));
-      try {
-        writeFileSync(join(tmp, '.gbrain-source'), 'essays\n');
-        const params = parseOpArgs(queryOp, ['find things']);
-        applyThinClientSourceScope(queryOp, params, tmp);
-        expect(params.source_id).toBe('essays');
-      } finally {
-        rmSync(tmp, { recursive: true, force: true });
-      }
-    });
-  });
-
-  test('explicit --source-id on the wire wins over ambient env scope', async () => {
-    await withEnv({ GBRAIN_SOURCE: 'gstack' }, () => {
-      const params = parseOpArgs(queryOp, ['find things', '--source-id', 'wiki']);
-      applyThinClientSourceScope(queryOp, params, '/');
-      expect(params.source_id).toBe('wiki');
-    });
-  });
-
-  test('--source together with --source-id is rejected loudly', async () => {
-    await withEnv({ GBRAIN_SOURCE: undefined }, () => {
-      const params = parseOpArgs(queryOp, ['q', '--source', 'a', '--source-id', 'b']);
-      expect(() => applyThinClientSourceScope(queryOp, params, '/')).toThrow(/not both/);
-    });
-  });
-
-  test('invalid --source value is rejected loudly', async () => {
-    await withEnv({ GBRAIN_SOURCE: undefined }, () => {
-      const params = parseOpArgs(queryOp, ['q', '--source', 'Bad_Value!']);
-      expect(() => applyThinClientSourceScope(queryOp, params, '/')).toThrow(/Invalid --source/);
-    });
-  });
-
-  test('--source on an op with no source_id wire param errors instead of silently dropping', async () => {
-    await withEnv({ GBRAIN_SOURCE: undefined }, () => {
-      const op = operationsByName.add_tag;
-      expect('source_id' in op.params).toBe(false);
-      const params = { slug: 'x', tag: 'y', source: 'wiki' };
-      expect(() => applyThinClientSourceScope(op, params, '/')).toThrow(/--source/);
-    });
-  });
-
-  test('ambient env scope on an op with no source_id wire param is ignored (no throw)', async () => {
-    await withEnv({ GBRAIN_SOURCE: 'wiki' }, () => {
-      const op = operationsByName.add_tag;
-      const params: Record<string, unknown> = { slug: 'x', tag: 'y' };
-      applyThinClientSourceScope(op, params, '/');
-      expect(params.source_id).toBeUndefined();
-    });
-  });
-
-  test('ops that declare their OWN source param are left untouched', async () => {
-    await withEnv({ GBRAIN_SOURCE: undefined }, () => {
-      const op = operationsByName.put_raw_data;
-      expect('source' in op.params).toBe(true);
-      const params: Record<string, unknown> = { slug: 'x', source: 'crustdata', data: {} };
-      applyThinClientSourceScope(op, params, '/');
-      expect(params.source).toBe('crustdata');
-      expect(params.source_id).toBeUndefined();
-    });
-  });
-
-  test('get_skill: ambient scope never leaks into its non-scope source_id param', async () => {
-    await withEnv({ GBRAIN_SOURCE: 'wiki' }, () => {
-      const op = operationsByName.get_skill;
-      expect('source_id' in op.params).toBe(true); // has the param, but it is a mode switch
-      const params: Record<string, unknown> = { name: 'ingest' };
-      applyThinClientSourceScope(op, params, '/');
-      expect(params.source_id).toBeUndefined(); // would flip host catalog → brain-pack lookup
-    });
-  });
-
-  test('get_skill: explicit --source errors instead of masquerading as --source-id', async () => {
-    await withEnv({ GBRAIN_SOURCE: undefined }, () => {
-      const op = operationsByName.get_skill;
-      const params: Record<string, unknown> = { name: 'ingest', source: 'wiki' };
-      expect(() => applyThinClientSourceScope(op, params, '/')).toThrow(/--source-id/);
-    });
-  });
-
-  test('get_skill: explicit --source-id passes through untouched', async () => {
-    await withEnv({ GBRAIN_SOURCE: 'gstack' }, () => {
-      const op = operationsByName.get_skill;
-      const params: Record<string, unknown> = { name: 'ingest', source_id: 'wiki' };
-      applyThinClientSourceScope(op, params, '/');
-      expect(params.source_id).toBe('wiki');
-    });
-  });
-
-  test('no scope from any tier leaves params unchanged', async () => {
-    await withEnv({ GBRAIN_SOURCE: undefined }, () => {
-      const params = parseOpArgs(queryOp, ['find things']);
-      applyThinClientSourceScope(queryOp, params, '/');
-      expect(params.source_id).toBeUndefined();
-    });
-  });
-});