fix(extract): resolve cross-directory wikilinks via sync-consistent slugification (#1964)

Raw Obsidian wikilink paths ([[llm-wiki/entities/AI 3.0]]) never matched
sync-slugified slugs (llm-wiki/entities/ai-3.0) on either extraction path.

- FS path: resolveSlug now slugifies each candidate (slugifyPath, the same
  function sync uses) when the raw path-equality check misses, in both the
  direct join and the ancestor walk. Dir-qualified wikilinks resolve as
  typed edges without the global_basename flag.
- DB path: extractPageLinks resolves dir-qualified generic wikilinks by
  slugifying the path and matching by exact slug or path suffix against
  the basename index — regardless of global_basename (the flag guards
  bare-name cross-dir ambiguity, which doesn't apply to dir-qualified
  paths). A same-tail page in a different directory can no longer be the
  only candidate and is filtered out.

Fixes #1964

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

src/commands/extract.ts

@@ -43,7 +43,7 @@ import {
 } from '../core/link-extraction.ts';
 import { createProgress } from '../core/progress.ts';
 import { getCliOptions, cliOptsToProgressOptions } from '../core/cli-options.ts';
-import { pathToSlug, pruneDir, isSyncable } from '../core/sync.ts';
+import { pathToSlug, slugifyPath, pruneDir, isSyncable } from '../core/sync.ts';
 // v0.41.18.0: withRetry + isRetryableConnError + WithRetryOpts moved to
 // src/core/retry.ts as the canonical primitive. Engine methods
 // (addLinksBatch/addTimelineEntriesBatch/upsertChunks) now self-retry via
@@ -269,14 +269,24 @@ export function extractMarkdownLinks(content: string): { name: string; relTarget
 export function resolveSlug(fileDir: string, relTarget: string, allSlugs: Set<string>): string | null {
   const targetNoExt = relTarget.endsWith('.md') ? relTarget.slice(0, -3) : relTarget;
 
-  const s1 = join(fileDir, targetNoExt);
-  if (allSlugs.has(s1)) return s1;
+  // Issue #1964: wikilinks carry raw Obsidian paths (`[[llm-wiki/entities/AI 3.0]]`)
+  // but allSlugs holds sync-slugified slugs (`llm-wiki/entities/ai-3.0`). Try the
+  // raw candidate first (back-compat), then the sync-consistent slugified form.
+  const hit = (candidate: string): string | null => {
+    if (allSlugs.has(candidate)) return candidate;
+    const slugified = slugifyPath(candidate);
+    if (slugified !== candidate && allSlugs.has(slugified)) return slugified;
+    return null;
+  };
+
+  const s1 = hit(join(fileDir, targetNoExt));
+  if (s1) return s1;
 
   const parts = fileDir.split('/').filter(Boolean);
   for (let strip = 1; strip <= parts.length; strip++) {
     const ancestor = parts.slice(0, parts.length - strip).join('/');
-    const candidate = ancestor ? join(ancestor, targetNoExt) : targetNoExt;
-    if (allSlugs.has(candidate)) return candidate;
+    const candidate = hit(ancestor ? join(ancestor, targetNoExt) : targetNoExt);
+    if (candidate) return candidate;
   }
 
   return null;

src/core/ai/capabilities.ts

@@ -22,7 +22,6 @@
  */
 
 import { resolveRecipe } from './model-resolver.ts';
-import { listRecipes } from './recipes/index.ts';
 import { AIConfigError } from './errors.ts';
 
 export interface ProviderCapabilities {
@@ -78,10 +77,7 @@ export function getProviderCapabilities(modelString: string): ProviderCapabiliti
   if (!chat) {
     throw new AIConfigError(
       `Provider "${recipe.id}" does not offer a chat touchpoint.`,
-      // 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.`,
+      `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.`,
     );
   }
 

src/core/ai/recipes/zhipu.ts

@@ -1,10 +1,9 @@
 import type { Recipe } from '../types.ts';
 
 /**
- * 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).
+ * 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).
  *
  * embedding-3 at 2048 dims exceeds pgvector's HNSW cap of 2000 — those
  * brains fall back to exact vector scans (see
@@ -26,20 +25,6 @@ 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,
@@ -51,5 +36,5 @@ export const zhipu: Recipe = {
     },
   },
   setup_hint:
-    'Get an API key at https://open.bigmodel.cn/, then `export ZHIPUAI_API_KEY=...`. Chat/subagent: use `zhipu:glm-5.1`.',
+    'Get an API key at https://open.bigmodel.cn/, then `export ZHIPUAI_API_KEY=...`',
 };

src/core/link-extraction.ts

@@ -14,6 +14,7 @@
 import type { BrainEngine } from './engine.ts';
 import type { PageType } from './types.ts';
 import { ensureWellFormed } from './text-safe.ts';
+import { slugifyPath } from './sync.ts';
 
 /**
  * v0.42.7 — link-extraction version stamp. Bump this ISO timestamp whenever the
@@ -482,14 +483,29 @@ export async function extractPageLinks(
     // pre-v0.40.8.2 behavior of dropping bare wikilinks outside
     // DIR_PATTERN.
     if (ref.needsResolution) {
-      if (!opts.globalBasename || typeof resolver.resolveBasenameMatches !== 'function') {
-        continue;
-      }
+      if (typeof resolver.resolveBasenameMatches !== 'function') continue;
       // Issue #972 (codex): resolve by the wikilink TARGET (ref.slug — the
       // text inside `[[...]]` before any `|`), NOT the display alias
       // (ref.name = match[2]). `[[struktura|the project]]` must resolve
       // `struktura`, not "the project". The display text is for context only.
-      const matches = await resolver.resolveBasenameMatches(ref.slug);
+      //
+      // Issue #1964: a dir-qualified wikilink (`[[llm-wiki/entities/AI 3.0]]`)
+      // carries a raw Obsidian path while page slugs are sync-slugified
+      // (`llm-wiki/entities/ai-3.0`). Slugify the path the same way sync does,
+      // then match by exact slug or path-suffix (wiki-root-relative authoring).
+      // This runs regardless of global_basename — it's dir-qualified, so the
+      // cross-dir false-positive risk the flag guards against doesn't apply.
+      // Mirrors the FS path's resolveSlug ancestor walk. Bare `[[name]]`
+      // wikilinks still require the global_basename flag.
+      let matches: string[] = [];
+      const slugified = ref.slug.includes('/') ? slugifyPath(ref.slug) : '';
+      if (slugified.includes('/')) {
+        const tail = slugified.slice(slugified.lastIndexOf('/') + 1);
+        matches = (await resolver.resolveBasenameMatches(tail))
+          .filter(m => m === slugified || m.endsWith(`/${slugified}`));
+      } else if (opts.globalBasename) {
+        matches = await resolver.resolveBasenameMatches(ref.slug);
+      }
       if (matches.length === 0) continue;
       const idx = content.indexOf(ref.slug);
       const context = idx >= 0 ? excerpt(content, idx, 240) : ref.name;

test/ai/recipe-zhipu.test.ts

@@ -69,45 +69,6 @@ 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/extract-fs.test.ts

@@ -389,6 +389,33 @@ describe('resolveSlugAll', () => {
   });
 });
 
+// ─── issue #1964: cross-directory wikilinks — slug/path mismatch ──────────
+
+describe('issue #1964: raw Obsidian wikilink paths resolve to sync-slugified slugs', () => {
+  test('resolveSlug slugifies the candidate (sync-consistent), no flag needed', () => {
+    const all = new Set(['llm-wiki/entities/ai-3.0']);
+    // Wikilink literal `[[llm-wiki/entities/AI 3.0]]` — spaces + uppercase.
+    expect(resolveSlug('llm-wiki/notes', 'llm-wiki/entities/AI 3.0.md', all))
+      .toBe('llm-wiki/entities/ai-3.0');
+  });
+
+  test('resolveSlug slugifies raw (unslugified) fileDir too', () => {
+    const all = new Set(['llm-wiki/entities/ai-3.0']);
+    // fileDir comes from dirname(relPath) — the raw on-disk directory.
+    expect(resolveSlug('LLM Wiki/Notes', 'entities/AI 3.0.md', all))
+      .toBe('llm-wiki/entities/ai-3.0');
+  });
+
+  test('extractLinksFromFile resolves cross-directory wikilink with flag OFF as a typed edge', async () => {
+    const allSlugs = new Set(['llm-wiki/entities/ai-3.0', 'llm-wiki/notes/roadmap']);
+    const content = '---\ntitle: Roadmap\ntype: concept\n---\n\nSee [[llm-wiki/entities/AI 3.0]].\n';
+    const links = await extractLinksFromFile(content, 'llm-wiki/notes/roadmap.md', allSlugs);
+    expect(links.map(l => l.to_slug)).toEqual(['llm-wiki/entities/ai-3.0']);
+    // Dir-qualified path resolution is exact, NOT the basename fallback.
+    expect(links[0].link_type).not.toBe('wikilink_basename');
+  });
+});
+
 describe('issue #972 repro: bare wikilinks resolve when flag is on', () => {
   // End-to-end: reproduces the issue's exact repro inside a tempdir +
   // PGLite, then asserts edge count under both flag states.

test/link-extraction.test.ts

@@ -9,6 +9,8 @@ import {
   parseTimelineEntries,
   isAutoLinkEnabled,
   FRONTMATTER_LINK_MAP,
+  buildBasenameIndex,
+  queryBasenameIndex,
   type SlugResolver,
 } from '../src/core/link-extraction.ts';
 import type { BrainEngine } from '../src/core/engine.ts';
@@ -424,6 +426,46 @@ describe('extractPageLinks', () => {
     expect(strk!.linkType).toBe('wikilink_basename');
   });
 
+  // ─── issue #1964: dir-qualified wikilinks with raw Obsidian paths ────────
+
+  test('#1964: dir-qualified wikilink resolves via sync-consistent slugification (flag OFF)', async () => {
+    // `[[llm-wiki/entities/AI 3.0]]` is a raw Obsidian path; the page slug
+    // is the sync-slugified `llm-wiki/entities/ai-3.0`. Must resolve WITHOUT
+    // global_basename (it's dir-qualified) and must NOT leak to a same-tail
+    // page in a different directory.
+    const resolver: SlugResolver = {
+      resolve: async () => null,
+      resolveBasenameMatches: async (name) =>
+        name === 'ai-3.0' ? ['other/ai-3.0', 'llm-wiki/entities/ai-3.0'] : [],
+    };
+    const { candidates } = await extractPageLinks(
+      'llm-wiki/notes/roadmap',
+      'See [[llm-wiki/entities/AI 3.0]] for the model.',
+      {}, 'concept', resolver,
+      // opts.globalBasename omitted (= false) — path is dir-qualified
+    );
+    expect(candidates.map(c => c.targetSlug)).toEqual(['llm-wiki/entities/ai-3.0']);
+    expect(candidates[0].linkType).toBe('wikilink_basename');
+    expect(candidates[0].linkSource).toBe('wikilink-resolved');
+  });
+
+  test('#1964: path-suffix match resolves wiki-root-relative paths against a real index', async () => {
+    // Author writes `[[llm-wiki/entities/AI 3.0]]` but the brain nests the
+    // wiki under a vault dir. Suffix match rescues it; queried through the
+    // REAL basename index so the tail-key lookup is exercised end to end.
+    const idx = buildBasenameIndex(['vault/llm-wiki/entities/ai-3.0', 'people/ai-3.0']);
+    const resolver: SlugResolver = {
+      resolve: async () => null,
+      resolveBasenameMatches: async (name) => queryBasenameIndex(idx, name),
+    };
+    const { candidates } = await extractPageLinks(
+      'vault/llm-wiki/notes/roadmap',
+      'See [[llm-wiki/entities/AI 3.0]].',
+      {}, 'concept', resolver,
+    );
+    expect(candidates.map(c => c.targetSlug)).toEqual(['vault/llm-wiki/entities/ai-3.0']);
+  });
+
   test('opts.skipFrontmatter suppresses the frontmatter pass', async () => {
     // Real resolver shape that WOULD resolve frontmatter source: too,
     // but skipFrontmatter blocks the path entirely.