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/cli.ts

@@ -935,10 +935,7 @@ export function formatResult(opName: string, result: unknown): string {
         lines.push(`Link coverage (entities): ${(h.link_coverage * 100).toFixed(1)}%`);
       }
       if (h.timeline_coverage !== undefined) {
-        lines.push(`Timeline coverage (entity pages): ${(h.timeline_coverage * 100).toFixed(1)}%`);
-      }
-      if (h.timeline_coverage_score !== undefined) {
-        lines.push(`Timeline density (all pages): ${h.timeline_coverage_score}/15 (whole-brain brain-score component)`);
+        lines.push(`Timeline coverage (entities): ${(h.timeline_coverage * 100).toFixed(1)}%`);
       }
       if (Array.isArray(h.most_connected) && h.most_connected.length > 0) {
         lines.push('Most connected entities:');

src/commands/doctor.ts

@@ -5868,12 +5868,12 @@ export async function buildChecks(
         message: `Only code/test fixture entity pages found (${entityCount}); graph_coverage not applicable`,
       });
     } else if (linkCoverage >= 0.5 && timelineCoverage >= 0.5) {
-      checks.push({ name: 'graph_coverage', status: 'ok', message: `Entity link coverage ${linkPct}%, entity timeline coverage ${timelinePct}%` });
+      checks.push({ name: 'graph_coverage', status: 'ok', message: `Entity link coverage ${linkPct}%, timeline ${timelinePct}%` });
     } else {
       checks.push({
         name: 'graph_coverage',
         status: 'warn',
-        message: `Entity link coverage ${linkPct}%, entity timeline coverage ${timelinePct}% (${eligibleEntityCount} entity pages). Run: gbrain extract all`,
+        message: `Entity link coverage ${linkPct}%, timeline ${timelinePct}% (${eligibleEntityCount} entity pages). Run: gbrain extract all`,
       });
     }
 
@@ -5885,7 +5885,7 @@ export async function buildChecks(
       const parts = [
         `embed ${health.embed_coverage_score}/35`,
         `links ${health.link_density_score}/25`,
-        `timeline density (all pages) ${health.timeline_coverage_score}/15`,
+        `timeline ${health.timeline_coverage_score}/15`,
         `orphans ${health.no_orphans_score}/15`,
         `dead-links ${health.no_dead_links_score}/10`,
       ];

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`.',
 };

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/doctor-timeline-metric-labels-2298.test.ts

@@ -1,155 +0,0 @@
-/**
- * Issue #2298 — timeline metric presentation contract.
- *
- * Authoritative upstream semantics (src/core/types.ts):
- *   - Metric A `timeline_coverage` (entity-scoped, fraction 0–1):
- *       eligible entity pages WITH a timeline entry / eligible entity pages
- *     -> surfaced by `graph_coverage` check AND `get_health` CLI entity line.
- *   - Metric B `timeline_coverage_score` (whole-brain, 0–15 brain-score component):
- *       all pages WITH a timeline entry / all pages
- *     -> surfaced by `brain_score` component breakdown AND (separately) CLI.
- *
- * The two have DIFFERENT numerators/denominators. This PR labels each
- * explicitly and keeps BOTH the entity CLI line and the whole-brain line.
- *
- * Tests (no private EriadorMu data, no production/home DB, no network):
- *   - numeric denominator assertions (Metric A = 50%, Metric B = 4/15)
- *   - doctor rendered-message assertions (exact labels, no ambiguous old label)
- *   - CLI rendered-output assertions (exact lines, guard matrix)
- *   - red/green: same assertions FAIL on origin/master, PASS on this branch
- *
- * Scoring formula UNCHANGED. Canonical PGLite fixture via resetPgliteState.
- */
-
-import { describe, expect, test, beforeAll, afterAll, beforeEach } from 'bun:test';
-import { PGLiteEngine } from '../src/core/pglite-engine.ts';
-import { sqlQueryForEngine } from '../src/core/sql-query.ts';
-import { resetPgliteState } from './helpers/reset-pglite.ts';
-import { buildChecks } from '../src/commands/doctor.ts';
-import { formatResult } from '../src/cli.ts';
-
-let engine: PGLiteEngine;
-
-async function seedFourPages(eng: PGLiteEngine): Promise<void> {
-  const sql = sqlQueryForEngine(eng);
-  // 2 eligible entity pages, 2 technical/non-entity pages.
-  // Only ONE entity page has a timeline entry; only ONE total page does.
-  await sql`
-    INSERT INTO pages (slug, source_id, type, title, compiled_truth, frontmatter, content_hash, created_at, updated_at)
-    VALUES
-      ('acme-example', 'default', 'company', 'Acme', '', '{}', 'h1', now(), now()),
-      ('alice-example', 'default', 'person', 'Alice', '', '{}', 'h2', now(), now()),
-      ('technical-a', 'default', 'note', 'Tech A', '', '{}', 'h3', now(), now()),
-      ('technical-b', 'default', 'note', 'Tech B', '', '{}', 'h4', now(), now())
-  `;
-  const companyId = (await sql`SELECT id FROM pages WHERE slug='acme-example'`)[0].id as number;
-  await sql`INSERT INTO timeline_entries (page_id, date, source, summary, detail)
-    VALUES (${companyId}, CURRENT_DATE, 'test', 'milestone', '{}')`;
-}
-
-beforeAll(async () => {
-  engine = new PGLiteEngine();
-  await engine.connect({});
-  await engine.initSchema();
-});
-
-afterAll(async () => {
-  await engine.disconnect();
-});
-
-beforeEach(async () => {
-  await resetPgliteState(engine);
-});
-
-describe('issue #2298 — numeric denominator semantics', () => {
-  test('entity timeline coverage = 1/2 = 50% (2 eligible entities, 1 with timeline)', async () => {
-    await seedFourPages(engine);
-    const health = await engine.getHealth();
-    expect(health.timeline_coverage).toBeDefined();
-    expect(Math.round((health.timeline_coverage ?? 0) * 100)).toBe(50);
-  });
-
-  test('whole-brain timeline density = 1/4 -> score 4/15 (4 total pages, 1 with timeline)', async () => {
-    await seedFourPages(engine);
-    const health = await engine.getHealth();
-    expect(health.timeline_coverage_score).toBeDefined();
-    expect(health.timeline_coverage_score).toBe(4);
-  });
-
-  test('the two metrics use independent denominators', async () => {
-    await seedFourPages(engine);
-    const health = await engine.getHealth();
-    expect(Math.round((health.timeline_coverage ?? 0) * 100)).toBe(50);
-    expect(health.timeline_coverage_score ?? 0).toBe(4);
-    // 50% (entity, /2) != 26.7% (whole-brain, /4). Provably distinct.
-    expect(Math.round(((health.timeline_coverage_score ?? 0) / 15) * 100)).not.toBe(50);
-  });
-});
-
-describe('issue #2298 — doctor rendered-message contract', () => {
-  test('graph_coverage renders entity-scoped label with 50%', async () => {
-    await seedFourPages(engine);
-    const checks = await buildChecks(engine, [], null);
-    const graph = checks.find((c) => c.name === 'graph_coverage');
-    expect(graph, 'graph_coverage check must be present').toBeDefined();
-    expect(graph!.message).toContain('entity timeline coverage 50%');
-    // ambiguous old label must NOT be present
-    expect(graph!.message).not.toMatch(/timeline 50%/);
-    expect(graph!.message).not.toMatch(/timeline \(entity, brain score\)/);
-  });
-
-  test('brain_score renders whole-brain density label 4/15', async () => {
-    await seedFourPages(engine);
-    const checks = await buildChecks(engine, [], null);
-    const brain = checks.find((c) => c.name === 'brain_score');
-    expect(brain, 'brain_score check must be present').toBeDefined();
-    expect(brain!.message).toContain('timeline density (all pages) 4/15');
-    // wrong labels must NOT be present
-    expect(brain!.message).not.toMatch(/timeline 4\/15/);
-    expect(brain!.message).not.toMatch(/timeline \(entity, brain score\)/);
-    // brain-score component must NOT carry the word "entity" (it is whole-brain)
-    const timelinePart = brain!.message.split('timeline density (all pages) 4/15')[0] + 'timeline density (all pages) 4/15';
-    expect(timelinePart).not.toMatch(/entity/);
-  });
-});
-
-describe('issue #2298 — CLI get_health rendered-output contract', () => {
-  function fakeHealth(overrides: Record<string, unknown>): any {
-    return {
-      embed_coverage: 1, missing_embeddings: 0, stale_pages: 0, orphan_pages: 0,
-      link_coverage: 1, timeline_coverage: 0.5, timeline_coverage_score: 4,
-      most_connected: [], ...overrides,
-    };
-  }
-
-  test('both entity and whole-brain lines render, no undefined/15', () => {
-    const out = formatResult('get_health', fakeHealth({}));
-    expect(out).toContain('Timeline coverage (entity pages): 50.0%');
-    expect(out).toContain('Timeline density (all pages): 4/15');
-    expect(out).not.toContain('undefined/15');
-    expect(out).not.toContain('Timeline coverage (entities)');
-    expect(out).not.toMatch(/timeline \(entity, brain score\)/);
-    expect(out).not.toMatch(/bare "timeline 4\/15"/);
-  });
-
-  test('guard matrix: entity present, whole-brain absent -> only entity line', () => {
-    const out = formatResult('get_health', fakeHealth({ timeline_coverage_score: undefined }));
-    expect(out).toContain('Timeline coverage (entity pages): 50.0%');
-    expect(out).not.toContain('Timeline density (all pages)');
-    expect(out).not.toContain('undefined/15');
-  });
-
-  test('guard matrix: whole-brain present, entity absent -> only whole-brain line', () => {
-    const out = formatResult('get_health', fakeHealth({ timeline_coverage: undefined }));
-    expect(out).toContain('Timeline density (all pages): 4/15');
-    expect(out).not.toContain('Timeline coverage (entity pages)');
-    expect(out).not.toContain('undefined/15');
-  });
-
-  test('guard matrix: both absent -> neither timeline line, never undefined/15', () => {
-    const out = formatResult('get_health', fakeHealth({ timeline_coverage: undefined, timeline_coverage_score: undefined }));
-    expect(out).not.toContain('Timeline coverage (entity pages)');
-    expect(out).not.toContain('Timeline density (all pages)');
-    expect(out).not.toContain('undefined/15');
-  });
-});