feat(lint): --exclude flag for mixed-content repos (takeover of #2649)

Adds --exclude=a,b (and LintOpts.exclude) so mixed-content repos can
skip software trees and repo metafiles by basename when collecting
pages. The only built-in default is node_modules — vendored dependency
trees are never knowledge pages; dot/underscore entries were already
skipped by the walk.

Diverges from #2649 deliberately: the original hardcoded an opinionated
default list (README.md, CHANGELOG.md, CLAUDE.md, test/ dirs at any
depth, plus fork-specific filenames), which silently changed lint
counts for every existing repo. Those are repo policy — pass --exclude.

Co-authored-by: ryangu00 <ryangu00@users.noreply.github.com>
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>

src/commands/lint.ts

@@ -378,15 +378,30 @@ async function resolveLintContentSanity(
   };
 }
 
+/**
+ * Directories never containing knowledge pages, skipped by default.
+ * Deliberately tiny: only vendored dependency trees qualify. Anything
+ * more opinionated (README.md, CHANGELOG.md, test/) is repo policy —
+ * callers opt in via `--exclude` / `LintOpts.exclude`. Dot- and
+ * underscore-prefixed entries are already skipped by the walk.
+ */
+const DEFAULT_LINT_EXCLUDE_DIRS = new Set(['node_modules']);
+
 /** Collect markdown files from a directory */
-function collectPages(dir: string): string[] {
+function collectPages(dir: string, extraExcludes: string[] = []): string[] {
+  const extra = new Set(extraExcludes);
   const pages: string[] = [];
   function walk(d: string) {
     for (const entry of readdirSync(d)) {
       if (entry.startsWith('.') || entry.startsWith('_')) continue;
       const full = join(d, entry);
-      if (lstatSync(full).isDirectory()) walk(full);
-      else if (entry.endsWith('.md')) pages.push(full);
+      if (lstatSync(full).isDirectory()) {
+        if (DEFAULT_LINT_EXCLUDE_DIRS.has(entry) || extra.has(entry)) continue;
+        walk(full);
+      } else if (entry.endsWith('.md')) {
+        if (extra.has(entry)) continue;
+        pages.push(full);
+      }
     }
   }
   walk(dir);
@@ -414,6 +429,13 @@ export interface LintOpts {
    * yields + checks this every 200 pages.
    */
   signal?: AbortSignal;
+  /**
+   * #2649: extra dir/file basenames to skip while collecting pages, in
+   * addition to node_modules and dot/underscore entries. For mixed-content
+   * repos (knowledge pages alongside software trees). Ignored for
+   * single-file targets.
+   */
+  exclude?: string[];
 }
 
 export interface LintResult {
@@ -440,7 +462,7 @@ export async function runLintCore(opts: LintOpts): Promise<LintResult> {
   }
 
   const isSingleFile = statSync(opts.target).isFile();
-  const pages = isSingleFile ? [opts.target] : collectPages(opts.target);
+  const pages = isSingleFile ? [opts.target] : collectPages(opts.target, opts.exclude ?? []);
 
   // Resolve content-sanity config once for this lint run (D1: lift DB
   // config when reachable). Caller can pre-pass via opts.contentSanity
@@ -491,14 +513,27 @@ export async function runLintCore(opts: LintOpts): Promise<LintResult> {
 }
 
 export async function runLint(args: string[]) {
-  const target = args.find(a => !a.startsWith('--'));
+  // #2649: --exclude=a,b or --exclude a,b — extra basenames to skip.
+  const extraExcludes: string[] = [];
+  const skipIdx = new Set<number>();
+  for (let i = 0; i < args.length; i++) {
+    const a = args[i];
+    if (a.startsWith('--exclude=')) {
+      extraExcludes.push(...a.slice('--exclude='.length).split(',').map(s => s.trim()).filter(Boolean));
+    } else if (a === '--exclude' && i + 1 < args.length) {
+      extraExcludes.push(...args[i + 1].split(',').map(s => s.trim()).filter(Boolean));
+      skipIdx.add(i + 1);
+    }
+  }
+  const target = args.find((a, i) => !a.startsWith('--') && !skipIdx.has(i));
   const doFix = args.includes('--fix');
   const dryRun = args.includes('--dry-run');
 
   if (!target) {
-    console.error('Usage: gbrain lint <dir|file.md> [--fix] [--dry-run]');
+    console.error('Usage: gbrain lint <dir|file.md> [--fix] [--dry-run] [--exclude a,b]');
     console.error('  --fix      Auto-fix fixable issues (LLM preambles, code fences)');
     console.error('  --dry-run  Preview fixes without writing');
+    console.error('  --exclude  Comma-separated dir/file basenames to skip (in addition to node_modules)');
     process.exit(1);
   }
 
@@ -510,7 +545,7 @@ export async function runLint(args: string[]) {
   // Single file or directory — print human detail as we go, then rely on
   // Core for the aggregate numbers at the end.
   const isSingleFile = statSync(target).isFile();
-  const pages = isSingleFile ? [target] : collectPages(target);
+  const pages = isSingleFile ? [target] : collectPages(target, extraExcludes);
 
   // Progress on stderr. Stdout keeps the per-issue human output it always had.
   const { createProgress } = await import('../core/progress.ts');
@@ -557,7 +592,7 @@ export async function runLint(args: string[]) {
   // produces canonical numbers for the summary line).
   // Pass contentSanity through so runLintCore skips its own resolve
   // (we already resolved once for the human-detail loop above).
-  const result = await runLintCore({ target, fix: doFix, dryRun, contentSanity });
+  const result = await runLintCore({ target, fix: doFix, dryRun, contentSanity, exclude: extraExcludes });
   console.log(`\n${result.pages_scanned} pages scanned. ${result.total_issues} issue(s) in ${result.pages_with_issues} page(s).`);
   if (doFix) {
     console.log(`${dryRun ? '(dry run) ' : ''}${result.total_fixed} auto-fixed.`);

test/lint.test.ts

@@ -116,3 +116,48 @@ describe('fixContent', () => {
     expect(fixed).toContain('# Title');
   });
 });
+
+describe('runLintCore exclude (takeover of #2649)', () => {
+  const { mkdtempSync, rmSync, mkdirSync, writeFileSync } = require('node:fs') as typeof import('node:fs');
+  const { tmpdir } = require('node:os') as typeof import('node:os');
+  const { join } = require('node:path') as typeof import('node:path');
+  const { runLintCore } = require('../src/commands/lint.ts') as typeof import('../src/commands/lint.ts');
+
+  const PAGE = '---\ntitle: T\ntype: note\ncreated: 2026-04-11\n---\n\n# T\n\nBody.\n';
+
+  function makeRepo(): string {
+    const dir = mkdtempSync(join(tmpdir(), 'gbrain-lint-excl-'));
+    writeFileSync(join(dir, 'page.md'), PAGE);
+    writeFileSync(join(dir, 'README.md'), PAGE);
+    mkdirSync(join(dir, 'node_modules', 'dep'), { recursive: true });
+    writeFileSync(join(dir, 'node_modules', 'dep', 'vendor.md'), PAGE);
+    mkdirSync(join(dir, 'software'));
+    writeFileSync(join(dir, 'software', 'notes.md'), PAGE);
+    return dir;
+  }
+
+  test('node_modules is excluded by default; nothing else is', async () => {
+    const dir = makeRepo();
+    try {
+      const result = await runLintCore({ target: dir, contentSanity: { disabled: true } });
+      // page.md + README.md + software/notes.md — vendor.md skipped.
+      expect(result.pages_scanned).toBe(3);
+    } finally {
+      rmSync(dir, { recursive: true, force: true });
+    }
+  });
+
+  test('--exclude basenames skip dirs and files', async () => {
+    const dir = makeRepo();
+    try {
+      const result = await runLintCore({
+        target: dir,
+        contentSanity: { disabled: true },
+        exclude: ['software', 'README.md'],
+      });
+      expect(result.pages_scanned).toBe(1); // only page.md
+    } finally {
+      rmSync(dir, { recursive: true, force: true });
+    }
+  });
+});