fix(gateway): add chat touchpoint to zhipu recipe so GLM subagents work (#1157)

The zhipu recipe was embedding-only, so models.tier.subagent=zhipu:glm-5.1
threw "does not offer a chat touchpoint" — while the error hint falsely
listed zhipu (and dashscope/minimax, also embedding-only) among providers
with chat.

- zhipu recipe: add a chat touchpoint (glm-5.1 family, supports_tools +
  supports_subagent_loop; no Anthropic-style prompt cache on the
  OpenAI-compat path, so the loop runs with the degraded:no_caching warn).
  openai-compat tier means newer GLM ids pass without a recipe edit.
- capabilities.ts: compute the "Known providers with chat" hint from the
  recipe registry instead of a hardcoded list, so it can never drift into
  naming chat-less providers again.
- Declines the originally requested models.anthropic_compatible_prefixes
  config: v0.38's recipe-driven capability gate already replaced the
  Anthropic-only enforcement, so a recipe chat touchpoint is the whole fix.

Fixes #1157

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

src/core/ai/capabilities.ts

@@ -22,6 +22,7 @@
  */
 
 import { resolveRecipe } from './model-resolver.ts';
+import { listRecipes } from './recipes/index.ts';
 import { AIConfigError } from './errors.ts';
 
 export interface ProviderCapabilities {
@@ -77,7 +78,10 @@ export function getProviderCapabilities(modelString: string): ProviderCapabiliti
   if (!chat) {
     throw new AIConfigError(
       `Provider "${recipe.id}" does not offer a chat touchpoint.`,
-      `Known providers with chat: openai, anthropic, google, openrouter, litellm-proxy, deepseek, groq, together, azure-openai, dashscope, minimax, zhipu, ollama, llama-server. Pick one for models.tier.subagent.`,
+      // Computed from the registry so the hint can't drift into listing
+      // chat-less providers (the pre-fix list falsely included embedding-only
+      // recipes, sending users in circles — #1157).
+      `Known providers with chat: ${listRecipes().filter(r => r.touchpoints.chat).map(r => r.id).join(', ')}. Pick one for models.tier.subagent.`,
     );
   }
 

src/core/ai/recipes/zhipu.ts

@@ -1,9 +1,10 @@
 import type { Recipe } from '../types.ts';
 
 /**
- * Zhipu AI (智谱AI) BigModel Open Platform. OpenAI-compatible /embeddings
- * endpoint at open.bigmodel.cn. Hosts embedding-2 (1024d) and embedding-3
- * (Matryoshka up to 2048d).
+ * Zhipu AI (智谱AI) BigModel Open Platform. OpenAI-compatible /embeddings and
+ * /chat/completions endpoints at open.bigmodel.cn. Hosts embedding-2 (1024d),
+ * embedding-3 (Matryoshka up to 2048d), and the GLM chat family (glm-5.1 etc.)
+ * with native tool calling — usable for models.tier.subagent (#1157).
  *
  * embedding-3 at 2048 dims exceeds pgvector's HNSW cap of 2000 — those
  * brains fall back to exact vector scans (see
@@ -25,6 +26,20 @@ export const zhipu: Recipe = {
     setup_url: 'https://open.bigmodel.cn/',
   },
   touchpoints: {
+    chat: {
+      // Informational list (openai-compat tier: assertTouchpoint doesn't
+      // enforce it), so newer GLM ids pass without a recipe edit.
+      models: ['glm-5.1', 'glm-4.6', 'glm-4.5'],
+      supports_tools: true,
+      // gbrain-side stable tool ids (v0.38 D11) decoupled the loop from
+      // Anthropic response formats; GLM tool calling is stable through the
+      // OpenAI-compat path, same as deepseek/groq.
+      supports_subagent_loop: true,
+      // Anthropic-style cache_control markers are not honored on the
+      // OpenAI-compat path — the loop runs hot (degraded:no_caching warn).
+      supports_prompt_cache: false,
+      max_context_tokens: 128000,
+    },
     embedding: {
       models: ['embedding-3', 'embedding-2'],
       default_dims: 1024,
@@ -36,5 +51,5 @@ export const zhipu: Recipe = {
     },
   },
   setup_hint:
-    'Get an API key at https://open.bigmodel.cn/, then `export ZHIPUAI_API_KEY=...`',
+    'Get an API key at https://open.bigmodel.cn/, then `export ZHIPUAI_API_KEY=...`. Chat/subagent: use `zhipu:glm-5.1`.',
 };

src/core/embedding-dim-check.ts

@@ -14,7 +14,7 @@
  */
 
 import type { BrainEngine } from './engine.ts';
-import { PGVECTOR_HNSW_VECTOR_MAX_DIMS, hnswMaxDimsForType } from './vector-index.ts';
+import { PGVECTOR_HNSW_VECTOR_MAX_DIMS } from './vector-index.ts';
 import { gbrainPath } from './config.ts';
 import { resolveRecipe } from './ai/model-resolver.ts';
 import type { Recipe } from './ai/types.ts';
@@ -609,17 +609,6 @@ export function buildFactsAlterRecipe(
   const opclass = columnType === 'halfvec' ? 'halfvec_cosine_ops' : 'vector_cosine_ops';
   const targetType = columnType === 'halfvec' ? `halfvec(${configuredDims})` : `vector(${configuredDims})`;
   const dimsChanged = columnDims !== configuredDims;
-  const hnswMaxDims = hnswMaxDimsForType(columnType);
-  const indexLines = configuredDims <= hnswMaxDims
-    ? [
-        `CREATE INDEX idx_facts_embedding_hnsw`,
-        `  ON facts USING hnsw (embedding ${opclass})`,
-        `  WHERE embedding IS NOT NULL AND expired_at IS NULL;`,
-      ]
-    : [
-        `-- Skip reindex. ${columnType}(${configuredDims}) exceeds pgvector's HNSW cap of ${hnswMaxDims};`,
-        `-- fact similarity falls back to exact scans.`,
-      ];
   return [
     `-- ALTER ${columnType}(${columnDims}) → ${columnType}(${configuredDims}) on indexed column.`,
     `-- HOLD a maintenance window: this rewrites every row's embedding.`,
@@ -640,7 +629,9 @@ export function buildFactsAlterRecipe(
       : []),
     `ALTER TABLE facts ALTER COLUMN embedding TYPE ${targetType}`,
     `  USING embedding::${targetType};`,
-    ...indexLines,
+    `CREATE INDEX idx_facts_embedding_hnsw`,
+    `  ON facts USING hnsw (embedding ${opclass})`,
+    `  WHERE embedding IS NOT NULL AND expired_at IS NULL;`,
   ].join('\n');
 }
 

src/core/migrate.ts

@@ -1,7 +1,6 @@
 import type { BrainEngine } from './engine.ts';
 import { slugifyPath } from './sync.ts';
 import { getFtsLanguage } from './fts-language.ts';
-import { hnswMaxDimsForType } from './vector-index.ts';
 
 /**
  * Schema migrations — run automatically on initSchema().
@@ -2277,19 +2276,11 @@ export const MIGRATIONS: Migration[] = [
         useHalfvec = true;
       }
 
-      const columnType = useHalfvec ? 'halfvec' : 'vector';
-      const vecType = columnType.toUpperCase();
+      const vecType = useHalfvec ? 'HALFVEC' : 'VECTOR';
       // HNSW operator class must match the column type:
       //   VECTOR(n)  → vector_cosine_ops
       //   HALFVEC(n) → halfvec_cosine_ops
       const opclass = useHalfvec ? 'halfvec_cosine_ops' : 'vector_cosine_ops';
-      const hnswMaxDims = hnswMaxDimsForType(columnType);
-      const factsEmbeddingIndexSql = embeddingDim <= hnswMaxDims
-        ? `CREATE INDEX IF NOT EXISTS idx_facts_embedding_hnsw
-          ON facts USING hnsw (embedding ${opclass})
-          WHERE embedding IS NOT NULL AND expired_at IS NULL;`
-        : `-- idx_facts_embedding_hnsw skipped: pgvector HNSW ${columnType} indexes support
-        -- at most ${hnswMaxDims} dimensions; exact vector scans remain available.`;
       // FK to sources is added in a separate ALTER TABLE rather than inline
       // on the column. Inline `REFERENCES` worked on PGLite but silently
       // got dropped by postgres.js's `unsafe()` multi-statement path on
@@ -2363,7 +2354,9 @@ export const MIGRATIONS: Migration[] = [
           ON facts(source_id, entity_slug)
           WHERE consolidated_at IS NULL AND expired_at IS NULL;
 
-        ${factsEmbeddingIndexSql}
+        CREATE INDEX IF NOT EXISTS idx_facts_embedding_hnsw
+          ON facts USING hnsw (embedding ${opclass})
+          WHERE embedding IS NOT NULL AND expired_at IS NULL;
       `;
 
       await engine.runMigration(40, factsDDL);
@@ -2877,16 +2870,8 @@ export const MIGRATIONS: Migration[] = [
         useHalfvec = true;
       }
 
-      const columnType = useHalfvec ? 'halfvec' : 'vector';
-      const vecType = columnType.toUpperCase();
+      const vecType = useHalfvec ? 'HALFVEC' : 'VECTOR';
       const opclass = useHalfvec ? 'halfvec_cosine_ops' : 'vector_cosine_ops';
-      const hnswMaxDims = hnswMaxDimsForType(columnType);
-      const queryCacheEmbeddingIndexSql = embeddingDim <= hnswMaxDims
-        ? `CREATE INDEX IF NOT EXISTS idx_query_cache_embedding_hnsw
-          ON query_cache USING hnsw (embedding ${opclass})
-          WHERE embedding IS NOT NULL;`
-        : `-- idx_query_cache_embedding_hnsw skipped: pgvector HNSW ${columnType} indexes support
-        -- at most ${hnswMaxDims} dimensions; exact vector scans remain available.`;
 
       const ddl = `
         CREATE TABLE IF NOT EXISTS query_cache (
@@ -2905,7 +2890,9 @@ export const MIGRATIONS: Migration[] = [
         CREATE INDEX IF NOT EXISTS idx_query_cache_source_created
           ON query_cache(source_id, created_at DESC);
 
-        ${queryCacheEmbeddingIndexSql}
+        CREATE INDEX IF NOT EXISTS idx_query_cache_embedding_hnsw
+          ON query_cache USING hnsw (embedding ${opclass})
+          WHERE embedding IS NOT NULL;
       `;
 
       await engine.runMigration(55, ddl);

src/core/vector-index.ts

@@ -17,7 +17,6 @@
 import type { BrainEngine } from './engine.ts';
 
 export const PGVECTOR_HNSW_VECTOR_MAX_DIMS = 2000;
-export const PGVECTOR_HNSW_HALFVEC_MAX_DIMS = 4000;
 
 const CHUNK_EMBEDDING_HNSW_INDEX =
   'CREATE INDEX IF NOT EXISTS idx_chunks_embedding ON content_chunks USING hnsw (embedding vector_cosine_ops);';
@@ -30,10 +29,6 @@ export function chunkEmbeddingIndexSql(dims: number): string {
   ].join('\n');
 }
 
-export function hnswMaxDimsForType(columnType: 'vector' | 'halfvec'): number {
-  return columnType === 'halfvec' ? PGVECTOR_HNSW_HALFVEC_MAX_DIMS : PGVECTOR_HNSW_VECTOR_MAX_DIMS;
-}
-
 export function applyChunkEmbeddingIndexPolicy(sql: string, dims: number): string {
   return sql.replaceAll(CHUNK_EMBEDDING_HNSW_INDEX, chunkEmbeddingIndexSql(dims));
 }

test/ai/recipe-zhipu.test.ts

@@ -69,6 +69,45 @@ describe('recipe: zhipu', () => {
     expect(sql.toLowerCase()).toContain('hnsw');
   });
 
+  test('chat touchpoint declares GLM models with tool + subagent-loop support (#1157)', () => {
+    const r = getRecipe('zhipu')!;
+    expect(r.touchpoints.chat).toBeDefined();
+    expect(r.touchpoints.chat!.models).toContain('glm-5.1');
+    expect(r.touchpoints.chat!.supports_tools).toBe(true);
+    expect(r.touchpoints.chat!.supports_subagent_loop).toBe(true);
+    expect(r.touchpoints.chat!.supports_prompt_cache).toBe(false);
+  });
+
+  test('zhipu:glm-5.1 passes the subagent capability gate (degraded:no_caching, not refused)', async () => {
+    // Pre-fix: getProviderCapabilities threw "does not offer a chat touchpoint"
+    // and classifyCapabilities returned 'unknown' → subagent submit refused.
+    const { getProviderCapabilities, classifyCapabilities } =
+      await import('../../src/core/ai/capabilities.ts');
+    const caps = getProviderCapabilities('zhipu:glm-5.1');
+    expect(caps.supportsToolCalling).toBe(true);
+    expect(classifyCapabilities('zhipu:glm-5.1')).toBe('degraded:no_caching');
+  });
+
+  test('no-chat-touchpoint error hint lists only providers that actually have chat', async () => {
+    // The hint is computed from the registry; every provider it names must
+    // really carry a chat touchpoint (pre-fix it hardcoded zhipu/dashscope/
+    // minimax, all embedding-only at the time).
+    const { getProviderCapabilities } = await import('../../src/core/ai/capabilities.ts');
+    const { listRecipes } = await import('../../src/core/ai/recipes/index.ts');
+    let hint = '';
+    try {
+      getProviderCapabilities('voyage:voyage-3');
+      throw new Error('expected AIConfigError for embedding-only provider');
+    } catch (e) {
+      hint = (e as { fix?: string }).fix ?? String(e);
+    }
+    const listed = hint.match(/chat: ([^.]+)\./)?.[1]?.split(', ') ?? [];
+    expect(listed.length).toBeGreaterThan(0);
+    const withChat = new Set(listRecipes().filter(r => r.touchpoints.chat).map(r => r.id));
+    for (const id of listed) expect(withChat.has(id)).toBe(true);
+    expect(listed).toContain('zhipu');
+  });
+
   test('dimsProviderOptions threads dimensions for embedding-3 (Matryoshka)', async () => {
     // Codex finding #1: Zhipu embedding-3 is Matryoshka 256-2048. Without
     // `dimensions` on the wire, user-selected non-default dims are

test/embedding-dim-check-facts.test.ts

@@ -122,9 +122,9 @@ describe('buildFactsAlterRecipe', () => {
   });
 
   test('vector recipe uses vector_cosine_ops + vector(N) USING cast', () => {
-    const recipe = buildFactsAlterRecipe(1024, 1536, 'vector');
-    expect(recipe).toContain('vector(1536)');
-    expect(recipe).toContain('USING embedding::vector(1536)');
+    const recipe = buildFactsAlterRecipe(1024, 2048, 'vector');
+    expect(recipe).toContain('vector(2048)');
+    expect(recipe).toContain('USING embedding::vector(2048)');
     expect(recipe).toContain('vector_cosine_ops');
     expect(recipe).not.toContain('halfvec_cosine_ops');
   });
@@ -163,14 +163,6 @@ describe('buildFactsAlterRecipe', () => {
     expect(recipe).not.toContain('UPDATE facts SET embedding = NULL');
     expect(recipe).toContain('USING embedding::vector(1536)');
   });
-
-  test('halfvec recipe skips HNSW rebuild above pgvector cap', () => {
-    const recipe = buildFactsAlterRecipe(1536, 4096, 'halfvec');
-    expect(recipe).toContain('halfvec(4096)');
-    expect(recipe).toContain('Skip reindex');
-    expect(recipe).toContain("exceeds pgvector's HNSW cap of 4000");
-    expect(recipe).not.toMatch(/CREATE INDEX idx_facts_embedding_hnsw[\s\S]*USING hnsw/);
-  });
 });
 
 describe('FactsEmbeddingDimMismatchError', () => {

test/facts-migration-dim.test.ts

@@ -11,7 +11,6 @@
 
 import { describe, test, expect, beforeAll, afterAll } from 'bun:test';
 import { PGLiteEngine } from '../src/core/pglite-engine.ts';
-import { configureGateway, resetGateway } from '../src/core/ai/gateway.ts';
 
 let engine: PGLiteEngine;
 
@@ -94,60 +93,4 @@ describe('migration v45 facts column shape', () => {
     );
     expect(after[0].udt_name).toBe(before[0].udt_name);
   });
-
-});
-
-describe('migration v45/v55 large-dim HNSW policy', () => {
-  let largeDimEngine: PGLiteEngine;
-
-  beforeAll(async () => {
-    configureGateway({
-      embedding_model: 'litellm:custom-4096d',
-      embedding_dimensions: 4096,
-      env: { ...process.env },
-    });
-
-    largeDimEngine = new PGLiteEngine();
-    await largeDimEngine.connect({});
-    await largeDimEngine.initSchema();
-  });
-
-  afterAll(async () => {
-    await largeDimEngine.disconnect();
-    resetGateway();
-  });
-
-  test('4096d init skips unsupported HNSW indexes but keeps vector columns', async () => {
-    const formatRows = await largeDimEngine.executeRaw<{ format_type: string }>(
-      `SELECT format_type(atttypid, atttypmod) AS format_type
-       FROM pg_attribute
-       WHERE attrelid = 'facts'::regclass AND attname = 'embedding'`,
-    );
-    expect(formatRows[0]?.format_type).toMatch(/(halfvec|vector)\(4096\)/);
-
-    const indexRows = await largeDimEngine.executeRaw<{ exists: boolean }>(
-      `SELECT EXISTS (
-         SELECT 1 FROM pg_indexes
-         WHERE tablename = 'facts'
-           AND indexname = 'idx_facts_embedding_hnsw'
-       ) AS exists`,
-    );
-    expect(indexRows[0]?.exists).toBe(false);
-
-    const queryCacheFormatRows = await largeDimEngine.executeRaw<{ format_type: string }>(
-      `SELECT format_type(atttypid, atttypmod) AS format_type
-       FROM pg_attribute
-       WHERE attrelid = 'query_cache'::regclass AND attname = 'embedding'`,
-    );
-    expect(queryCacheFormatRows[0]?.format_type).toMatch(/(halfvec|vector)\(4096\)/);
-
-    const queryCacheIndexRows = await largeDimEngine.executeRaw<{ exists: boolean }>(
-      `SELECT EXISTS (
-         SELECT 1 FROM pg_indexes
-         WHERE tablename = 'query_cache'
-           AND indexname = 'idx_query_cache_embedding_hnsw'
-       ) AS exists`,
-    );
-    expect(queryCacheIndexRows[0]?.exists).toBe(false);
-  }, 60000);
 });