fix(slugs): CJK slug support in SlugRegistry and dream-cycle summary slug (takeover of #782, #738)

Master already widened slugifySegment (sync.ts) and validatePageSlug
(operations.ts) to CJK in v0.32.7, but the other two validators #782
targeted stayed ASCII-only: SlugRegistry's SLUG_RE rejected any CJK
desiredSlug from BrainWriter, and synthesize.ts's SUMMARY_SLUG_RE (whose
comment claimed it was kept in sync with validatePageSlug) rejected CJK
output roots.

Hoist the segment grammar into cjk.ts as PAGE_SLUG_SEG and compose all
three regex sites from it, so the four slug validators share one grammar.
Each site keeps its own shape (SlugRegistry's >=2-segment dir/name form,
validatePageSlug's case-insensitive flag).

Scope stays CJK (matching v0.32.7), not full \p{L} Unicode as #782
proposed — all-scripts slugs (lookalike/RTL spoofing) is a maintainer
policy call.

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

src/commands/autopilot.ts

@@ -151,100 +151,6 @@ export function shouldSpawnAutopilotWorker(args: string[]): boolean {
   return !args.includes('--no-worker');
 }
 
-/**
- * #1525 — positional subcommand translation.
- *
- * Pre-fix, `gbrain autopilot status` silently fell through to "start daemon"
- * because `runAutopilot()` only branched on flag forms (`--status`, etc.).
- * `status` was treated as a stray positional and ignored.
- *
- * This translator maps known positional subcommands to their flag form so
- * `autopilot status` is equivalent to `autopilot --status`, then rejects
- * any unrecognized positional with a fail-loud error before any side
- * effect (lockfile, daemon spawn, sync dispatch) runs.
- *
- * Scope decisions:
- *   - Known aliases: `status` → `--status`, `install` → `--install`,
- *     `uninstall` → `--uninstall`, `start` → (drop; default daemon launch).
- *   - `stop` is intentionally NOT aliased here. Stopping a running daemon
- *     is a new behavior (read PID from lock, SIGTERM, drain) that deserves
- *     its own design and PR. Users typing `gbrain autopilot stop` today get
- *     the unknown-positional error with the canonical alternatives.
- *   - At most one positional allowed; multiple positionals fail loud.
- */
-// Every flag that consumes the NEXT argv token. Missing one here makes the
-// translator misread the flag's value as a positional subcommand and exit 2
-// (e.g. `--install --target linux-cron`). Keep in sync with parseArg call sites.
-const AUTOPILOT_VALUE_FLAGS = new Set(['--repo', '--interval', '--target']);
-const AUTOPILOT_POSITIONAL_ALIASES: Record<string, string | null> = {
-  status: '--status',
-  install: '--install',
-  uninstall: '--uninstall',
-  start: null, // drop the positional; default behavior is daemon launch
-};
-
-export type PositionalTranslation =
-  | { ok: true; args: string[] }
-  | {
-      ok: false;
-      reason: 'unknown_subcommand' | 'multiple_subcommands';
-      message: string;
-    };
-
-export function translatePositionalSubcommands(args: string[]): PositionalTranslation {
-  const out: string[] = [];
-  let positionalSeen = false;
-  let i = 0;
-  while (i < args.length) {
-    const a = args[i];
-    if (AUTOPILOT_VALUE_FLAGS.has(a)) {
-      // Pass through the flag and its value untouched. If the value is
-      // missing at end-of-argv, fall through so the existing parseArg
-      // path can report the broken usage.
-      out.push(a);
-      if (i + 1 < args.length) {
-        out.push(args[i + 1]);
-        i += 2;
-      } else {
-        i += 1;
-      }
-      continue;
-    }
-    if (a.startsWith('-')) {
-      out.push(a);
-      i += 1;
-      continue;
-    }
-    // Positional subcommand.
-    if (positionalSeen) {
-      const known = Object.keys(AUTOPILOT_POSITIONAL_ALIASES).join(', ');
-      return {
-        ok: false,
-        reason: 'multiple_subcommands',
-        message: `Multiple subcommands given. Use only one of: ${known}.`,
-      };
-    }
-    positionalSeen = true;
-    if (a in AUTOPILOT_POSITIONAL_ALIASES) {
-      const alias = AUTOPILOT_POSITIONAL_ALIASES[a];
-      if (alias) out.push(alias);
-      i += 1;
-      continue;
-    }
-    const known = Object.keys(AUTOPILOT_POSITIONAL_ALIASES).join(', ');
-    return {
-      ok: false,
-      reason: 'unknown_subcommand',
-      message:
-        `Unknown subcommand: \`${a}\`.\n` +
-        `Allowed subcommands: ${known}.\n` +
-        `Or use the flag form: --status, --install, --uninstall.\n` +
-        `Run \`gbrain autopilot --help\` for full usage.`,
-    };
-  }
-  return { ok: true, args: out };
-}
-
 export function isPidAlive(pid: number): boolean {
   if (!Number.isFinite(pid) || pid <= 0) return false;
   try {
@@ -457,11 +363,6 @@ export async function runAutopilot(engine: BrainEngine, args: string[]) {
       '       gbrain autopilot --install [--repo <path>]\n' +
       '       gbrain autopilot --uninstall\n' +
       '       gbrain autopilot --status [--json]\n\n' +
-      'Subcommand aliases:\n' +
-      '       gbrain autopilot status     → --status\n' +
-      '       gbrain autopilot install    → --install\n' +
-      '       gbrain autopilot uninstall  → --uninstall\n' +
-      '       gbrain autopilot start      → (default daemon launch)\n\n' +
       'Self-maintaining brain daemon. Runs the full maintenance cycle\n' +
       '(lint + backlinks + sync + extract + embed + orphans) on an interval.\n\n' +
       'For a one-shot cron-triggered cycle, see `gbrain dream`.',
@@ -469,16 +370,6 @@ export async function runAutopilot(engine: BrainEngine, args: string[]) {
     return;
   }
 
-  // #1525: translate positional subcommands to their flag form BEFORE any
-  // side effect (lockfile, daemon spawn, sync dispatch). Unknown positionals
-  // fail loud here rather than silently starting the daemon.
-  const translated = translatePositionalSubcommands(args);
-  if (!translated.ok) {
-    console.error(translated.message);
-    process.exit(2);
-  }
-  args = translated.args;
-
   if (args.includes('--install')) {
     await installDaemon(engine, args);
     return;

src/core/cjk.ts

@@ -20,6 +20,14 @@ export const CJK_SLUG_CHARS = '一-鿿぀-ゟ゠-ヿ가-힯';
 
 export const CJK_RANGES_REGEX = new RegExp(`[${CJK_SLUG_CHARS}]`);
 
+/**
+ * Page-slug segment grammar (no anchors): alnum-or-CJK lead char, then
+ * alnum/CJK/hyphen continuation. Single source for validatePageSlug
+ * (operations.ts), SlugRegistry's SLUG_RE, and the dream-cycle
+ * SUMMARY_SLUG_RE so every slug validator shares one grammar (#738).
+ */
+export const PAGE_SLUG_SEG = `[a-z0-9${CJK_SLUG_CHARS}][a-z0-9${CJK_SLUG_CHARS}\\-]*`;
+
 export const CJK_SENTENCE_DELIMITERS = ['。', '！', '？']; // 。！？
 export const CJK_CLAUSE_DELIMITERS = ['；', '：', '，', '、']; // ；：，、
 

src/core/cycle/synthesize.ts

@@ -43,10 +43,11 @@ import { serializeMarkdown, serializePageToMarkdown } from '../markdown.ts';
 import type { Page, PageType } from '../types.ts';
 import { validateSourceId } from '../utils.ts';
 import { safeSplitIndex } from '../text-safe.ts';
+import { PAGE_SLUG_SEG } from '../cjk.ts';
 
-// Slug regex from validatePageSlug — kept in sync.
+// Slug grammar from validatePageSlug — shared via PAGE_SLUG_SEG (#738).
 // Used for the orchestrator-written summary index slug.
-const SUMMARY_SLUG_RE = /^[a-z0-9][a-z0-9\-]*(\/[a-z0-9][a-z0-9\-]*)*$/;
+const SUMMARY_SLUG_RE = new RegExp(`^${PAGE_SLUG_SEG}(\\/${PAGE_SLUG_SEG})*$`);
 
 // ── Model context budget (D1, D5, D7, D9) ─────────────────────────────
 

src/core/operations.ts

@@ -25,7 +25,7 @@ import { bumpLastRetrievedAt } from './last-retrieved.ts';
 import { isSearchMode } from './search/mode.ts';
 import { stampEvidence } from './search/evidence.ts';
 import type { SearchResult } from './types.ts';
-import { CJK_SLUG_CHARS } from './cjk.ts';
+import { CJK_SLUG_CHARS, PAGE_SLUG_SEG } from './cjk.ts';
 import * as db from './db.ts';
 import { VERSION } from '../version.ts';
 import {
@@ -162,7 +162,6 @@ export function validatePageSlug(slug: string): void {
   }
   // v0.32.7: CJK ranges (Han / Hiragana / Katakana / Hangul Syllables) allowed
   // in segments. ASCII shape rules (lead char, hyphen continuation) preserved.
-  const PAGE_SLUG_SEG = `[a-z0-9${CJK_SLUG_CHARS}][a-z0-9${CJK_SLUG_CHARS}\\-]*`;
   if (!new RegExp(`^${PAGE_SLUG_SEG}(\\/${PAGE_SLUG_SEG})*$`, 'i').test(slug)) {
     throw new OperationError('invalid_params', `Invalid page_slug: ${slug} (allowed: alphanumeric, CJK, hyphens, forward-slash separated segments)`);
   }

src/core/output/slug-registry.ts

@@ -17,6 +17,7 @@
 
 import type { BrainEngine } from '../engine.ts';
 import type { PageType } from '../types.ts';
+import { PAGE_SLUG_SEG } from '../cjk.ts';
 
 export interface CreateSlugInput {
   /**
@@ -71,7 +72,9 @@ export class SlugRegistryError extends Error {
 // SlugRegistry
 // ---------------------------------------------------------------------------
 
-const SLUG_RE = /^[a-z0-9][a-z0-9\-]*(\/[a-z0-9][a-z0-9\-]*)+$/;
+// Shares the page-slug segment grammar (incl. CJK ranges, #738) with
+// validatePageSlug; keeps this site's dir/name shape (>= 2 segments).
+const SLUG_RE = new RegExp(`^${PAGE_SLUG_SEG}(\\/${PAGE_SLUG_SEG})+$`);
 
 export class SlugRegistry {
   constructor(private engine: BrainEngine) {}

test/autopilot-positional-subcommands.test.ts

@@ -1,197 +0,0 @@
-/**
- * Tests for translatePositionalSubcommands() — the v0.41.x #1525 fix that
- * prevents `gbrain autopilot status` from silently starting the daemon.
- *
- * IRON RULE regression guard: the exact ticket repro (`gbrain autopilot
- * status`) MUST translate to `--status`, not fall through to the default
- * daemon launch. Verified by the "ticket-exact repro" case below.
- */
-
-import { describe, test, expect } from 'bun:test';
-import { translatePositionalSubcommands } from '../src/commands/autopilot.ts';
-
-describe('translatePositionalSubcommands — known aliases', () => {
-  test('IRON RULE — `autopilot status` translates to `--status` (ticket #1525 repro)', () => {
-    const r = translatePositionalSubcommands(['status']);
-    expect(r.ok).toBe(true);
-    if (r.ok) expect(r.args).toEqual(['--status']);
-  });
-
-  test('`install` translates to `--install`', () => {
-    const r = translatePositionalSubcommands(['install']);
-    expect(r.ok).toBe(true);
-    if (r.ok) expect(r.args).toEqual(['--install']);
-  });
-
-  test('`uninstall` translates to `--uninstall`', () => {
-    const r = translatePositionalSubcommands(['uninstall']);
-    expect(r.ok).toBe(true);
-    if (r.ok) expect(r.args).toEqual(['--uninstall']);
-  });
-
-  test('`start` drops the positional (default daemon launch)', () => {
-    const r = translatePositionalSubcommands(['start']);
-    expect(r.ok).toBe(true);
-    if (r.ok) expect(r.args).toEqual([]);
-  });
-
-  test('`start --json` drops only the positional, keeps the flag', () => {
-    const r = translatePositionalSubcommands(['start', '--json']);
-    expect(r.ok).toBe(true);
-    if (r.ok) expect(r.args).toEqual(['--json']);
-  });
-});
-
-describe('translatePositionalSubcommands — flag/positional interleaving', () => {
-  test('`status --json` preserves the trailing flag', () => {
-    const r = translatePositionalSubcommands(['status', '--json']);
-    expect(r.ok).toBe(true);
-    if (r.ok) expect(r.args).toEqual(['--status', '--json']);
-  });
-
-  test('`--json status` preserves the leading flag', () => {
-    const r = translatePositionalSubcommands(['--json', 'status']);
-    expect(r.ok).toBe(true);
-    if (r.ok) expect(r.args).toEqual(['--json', '--status']);
-  });
-
-  test('`--repo /foo status` does not mis-classify the path as positional', () => {
-    const r = translatePositionalSubcommands(['--repo', '/foo', 'status']);
-    expect(r.ok).toBe(true);
-    if (r.ok) expect(r.args).toEqual(['--repo', '/foo', '--status']);
-  });
-
-  test('`--interval 300 install` does not mis-classify the number as positional', () => {
-    const r = translatePositionalSubcommands(['--interval', '300', 'install']);
-    expect(r.ok).toBe(true);
-    if (r.ok) expect(r.args).toEqual(['--interval', '300', '--install']);
-  });
-
-  test('`--install --target linux-cron` does not mis-classify the target as positional', () => {
-    // --target is installDaemon's value flag; its value must never be read
-    // as a positional subcommand (regression guard for the review fix).
-    const r = translatePositionalSubcommands(['--install', '--target', 'linux-cron']);
-    expect(r.ok).toBe(true);
-    if (r.ok) expect(r.args).toEqual(['--install', '--target', 'linux-cron']);
-  });
-
-  test('`install --target macos` keeps the alias translation and the target value', () => {
-    const r = translatePositionalSubcommands(['install', '--target', 'macos']);
-    expect(r.ok).toBe(true);
-    if (r.ok) expect(r.args).toEqual(['--install', '--target', 'macos']);
-  });
-
-  test('value-flag at end of argv with missing value passes through (so parseArg can report it)', () => {
-    const r = translatePositionalSubcommands(['--repo']);
-    expect(r.ok).toBe(true);
-    if (r.ok) expect(r.args).toEqual(['--repo']);
-  });
-
-  test('value-flag whose value looks like an alias is NOT translated', () => {
-    // `--repo status` means "use repo path 'status'", not "show status".
-    // Translator must not destructure the value of --repo.
-    const r = translatePositionalSubcommands(['--repo', 'status']);
-    expect(r.ok).toBe(true);
-    if (r.ok) expect(r.args).toEqual(['--repo', 'status']);
-  });
-});
-
-describe('translatePositionalSubcommands — pass-through cases', () => {
-  test('empty args returns empty args', () => {
-    const r = translatePositionalSubcommands([]);
-    expect(r.ok).toBe(true);
-    if (r.ok) expect(r.args).toEqual([]);
-  });
-
-  test('flag-only invocation passes through unchanged', () => {
-    const r = translatePositionalSubcommands(['--status', '--json']);
-    expect(r.ok).toBe(true);
-    if (r.ok) expect(r.args).toEqual(['--status', '--json']);
-  });
-
-  test('short flag `-h` passes through unchanged', () => {
-    const r = translatePositionalSubcommands(['-h']);
-    expect(r.ok).toBe(true);
-    if (r.ok) expect(r.args).toEqual(['-h']);
-  });
-
-  test('all known bare flags pass through unchanged', () => {
-    const flags = ['--help', '--install', '--uninstall', '--status', '--json', '--inline', '--no-worker'];
-    const r = translatePositionalSubcommands(flags);
-    expect(r.ok).toBe(true);
-    if (r.ok) expect(r.args).toEqual(flags);
-  });
-});
-
-describe('translatePositionalSubcommands — rejection of unknown positionals', () => {
-  test('unknown positional `foo` fails with reason=unknown_subcommand + structured message', () => {
-    const r = translatePositionalSubcommands(['foo']);
-    expect(r.ok).toBe(false);
-    if (!r.ok) {
-      expect(r.reason).toBe('unknown_subcommand');
-      expect(r.message).toContain('Unknown subcommand: `foo`');
-      expect(r.message).toContain('status');
-      expect(r.message).toContain('install');
-      expect(r.message).toContain('uninstall');
-      expect(r.message).toContain('--help');
-    }
-  });
-
-  test('unknown positional `stop` fails with reason=unknown_subcommand (NOT silently aliased)', () => {
-    // Stop is mentioned in the ticket but deliberately NOT aliased in this
-    // PR — stopping a running daemon is a new behavior, not just an alias.
-    // Until that feature lands separately, `stop` must fail loud rather
-    // than starting the daemon (the bug we're fixing).
-    const r = translatePositionalSubcommands(['stop']);
-    expect(r.ok).toBe(false);
-    if (!r.ok) {
-      expect(r.reason).toBe('unknown_subcommand');
-      expect(r.message).toContain('Unknown subcommand: `stop`');
-    }
-  });
-
-  test('unknown positional `status-detail` (close-but-not-matching) fails', () => {
-    const r = translatePositionalSubcommands(['status-detail']);
-    expect(r.ok).toBe(false);
-    if (!r.ok) {
-      expect(r.reason).toBe('unknown_subcommand');
-      expect(r.message).toContain('Unknown subcommand: `status-detail`');
-    }
-  });
-
-  test('multiple positionals fail with reason=multiple_subcommands (`start install`)', () => {
-    const r = translatePositionalSubcommands(['start', 'install']);
-    expect(r.ok).toBe(false);
-    if (!r.ok) {
-      expect(r.reason).toBe('multiple_subcommands');
-      expect(r.message).toContain('Multiple subcommands');
-    }
-  });
-
-  test('multiple positionals fail even when both are known aliases (`status install`)', () => {
-    const r = translatePositionalSubcommands(['status', 'install']);
-    expect(r.ok).toBe(false);
-    if (!r.ok) {
-      expect(r.reason).toBe('multiple_subcommands');
-      expect(r.message).toContain('Multiple subcommands');
-    }
-  });
-
-  test('known-then-unknown rejects with multiple_subcommands (first-positional-wins)', () => {
-    // First positional is known, second is not. Rejection comes from the
-    // multiple-positional rule, which fires before the unknown check; the
-    // intent is "only one subcommand allowed."
-    const r = translatePositionalSubcommands(['status', 'garbage']);
-    expect(r.ok).toBe(false);
-    if (!r.ok) expect(r.reason).toBe('multiple_subcommands');
-  });
-
-  test('unknown-then-known rejects on the unknown (unknown fires before second-positional check)', () => {
-    const r = translatePositionalSubcommands(['garbage', 'status']);
-    expect(r.ok).toBe(false);
-    if (!r.ok) {
-      expect(r.reason).toBe('unknown_subcommand');
-      expect(r.message).toContain('garbage');
-    }
-  });
-});

test/cycle-dream-output-root.test.ts

@@ -86,6 +86,12 @@ describe('#2415: loadOutputRoot validation + patterns gather scope', () => {
     expect(await loadOutputRoot(engine)).toBe('wiki');
   });
 
+  test('CJK root passes the slug grammar (#738)', async () => {
+    await engine.setConfig('dream.synthesize.output_root', '知识/笔记');
+    expect(await loadOutputRoot(engine)).toBe('知识/笔记');
+    await engine.setConfig('dream.synthesize.output_root', '');
+  });
+
   test('patterns phase gathers reflections under the configured root', async () => {
     await engine.setConfig('dream.synthesize.output_root', 'notes');
     for (let i = 0; i < 3; i++) {

test/writer.test.ts

@@ -203,6 +203,17 @@ describe('SlugRegistry', () => {
     })).rejects.toThrow(SlugRegistryError);
   });
 
+  test('create accepts CJK slugs (#738)', async () => {
+    const reg = new SlugRegistry(engine);
+    const r = await reg.create({
+      desiredSlug: '知识/品牌圣经',
+      displayName: '品牌圣经',
+      type: 'note',
+    });
+    expect(r.slug).toBe('知识/品牌圣经');
+    expect(r.exact).toBe(true);
+  });
+
   test('create throws on invalid slug', async () => {
     const reg = new SlugRegistry(engine);
     await expect(reg.create({