fix(shell): use git ls-files for @ file mention in large repos

CHANGELOG.md

@@ -11,6 +11,10 @@ Only write entries that are worth mentioning to users.
 
 ## Unreleased
 
+- Shell: Use `git ls-files` for `@` file mention discovery — file completer now queries `git ls-files --recurse-submodules` with a 5-second timeout as the primary discovery mechanism, falling back to `os.walk` for non-git repositories; this fixes large repositories (e.g., apache/superset with 65k+ files) where the 1000-file limit caused late-alphabetical directories to be unreachable (fixes #1375)
+- Core: Add shared `file_filter` module — unifies file mention logic between shell and web UIs via `src/kimi_cli/utils/file_filter.py`, providing consistent path filtering, ignored directory exclusion, and git-aware file discovery
+- Shell: Prevent path traversal in file mention scope parameter — the `scope` parameter in file completer requests is now validated to prevent directory traversal attacks
+- Web: Restore unfiltered directory listing in file browser API — file browser endpoint no longer applies git-aware filtering, ensuring all files are visible in the web UI file picker
 - Todo: Refactor SetTodoList to persist state and prevent tool call storms — todos are now persisted to session state (root agent) and independent state files (sub-agents); adds query mode (omit `todos` to read current state) and clear mode (pass `[]`); includes anti-storm guidance in tool description to prevent repeated calls without progress (fixes #1710)
 - ReadFile: Add total line count to every read response and support negative `line_offset` for tail mode — the tool now reports `Total lines in file: N.` in its message so the model can plan subsequent reads; negative `line_offset` (e.g. `-100`) reads the last N lines using a sliding window, useful for viewing recent log output without shell commands; the absolute value is capped at 1000 (MAX_LINES)
 - Shell: Fix black background on inline code and code blocks in Markdown rendering — `NEUTRAL_MARKDOWN_THEME` now overrides all Rich default `markdown.*` styles to `"none"`, preventing Rich's built-in `"cyan on black"` from leaking through on non-black terminals

docs/en/release-notes/changelog.md

@@ -4,6 +4,10 @@ This page documents the changes in each Kimi Code CLI release.
 
 ## Unreleased
 
+- Shell: Use `git ls-files` for `@` file mention discovery — file completer now queries `git ls-files --recurse-submodules` with a 5-second timeout as the primary discovery mechanism, falling back to `os.walk` for non-git repositories; this fixes large repositories (e.g., apache/superset with 65k+ files) where the 1000-file limit caused late-alphabetical directories to be unreachable (fixes #1375)
+- Core: Add shared `file_filter` module — unifies file mention logic between shell and web UIs via `src/kimi_cli/utils/file_filter.py`, providing consistent path filtering, ignored directory exclusion, and git-aware file discovery
+- Shell: Prevent path traversal in file mention scope parameter — the `scope` parameter in file completer requests is now validated to prevent directory traversal attacks
+- Web: Restore unfiltered directory listing in file browser API — file browser endpoint no longer applies git-aware filtering, ensuring all files are visible in the web UI file picker
 - Todo: Refactor SetTodoList to persist state and prevent tool call storms — todos are now persisted to session state (root agent) and independent state files (sub-agents); adds query mode (omit `todos` to read current state) and clear mode (pass `[]`); includes anti-storm guidance in tool description to prevent repeated calls without progress (fixes #1710)
 - ReadFile: Add total line count to every read response and support negative `line_offset` for tail mode — the tool now reports `Total lines in file: N.` in its message so the model can plan subsequent reads; negative `line_offset` (e.g. `-100`) reads the last N lines using a sliding window, useful for viewing recent log output without shell commands; the absolute value is capped at 1000 (MAX_LINES)
 - Shell: Fix black background on inline code and code blocks in Markdown rendering — `NEUTRAL_MARKDOWN_THEME` now overrides all Rich default `markdown.*` styles to `"none"`, preventing Rich's built-in `"cyan on black"` from leaking through on non-black terminals

docs/zh/release-notes/changelog.md

@@ -4,6 +4,10 @@
 
 ## 未发布
 
+- Shell：使用 `git ls-files` 进行 `@` 文件引用发现——文件补全器现在优先使用 `git ls-files --recurse-submodules` 查询文件列表（5 秒超时），非 Git 仓库则回退到 `os.walk`；此修复解决了大型仓库（如包含 6.5 万+文件的 apache/superset）中 1000 文件限制导致字母顺序靠后的目录无法访问的问题（修复 #1375）
+- Core：新增共享的 `file_filter` 模块——通过 `src/kimi_cli/utils/file_filter.py` 统一 Shell 和 Web 的文件引用逻辑，提供一致的路径过滤、忽略目录排除和 Git 感知文件发现
+- Shell：防止文件引用 scope 参数的路径遍历——文件补全器请求中的 `scope` 参数现在会经过验证，防止目录遍历攻击
+- Web：恢复文件浏览器 API 中的未过滤目录列表——文件浏览器端点不再应用 Git 感知过滤，确保 Web UI 文件选择器中显示所有文件
 - Todo：重构 `SetTodoList` 工具，支持状态持久化并防止工具调用风暴——待办事项现在会持久化到会话状态（主 Agent）和独立状态文件（子 Agent）；新增查询模式（省略 `todos` 参数可读取当前状态）和清空模式（传 `[]` 清空）；工具描述中增加了防风暴指导，防止在没有实际进展的情况下反复调用（修复 #1710）
 - ReadFile：每次读取返回文件总行数，并支持负数 `line_offset` 实现 tail 模式——工具现在会在消息中报告 `Total lines in file: N.`，方便模型规划后续读取；负数 `line_offset`（如 `-100`）通过滑动窗口读取文件末尾 N 行，适用于无需 Shell 命令即可查看最新日志输出的场景；绝对值上限为 1000（MAX_LINES）
 - Shell：修复 Markdown 渲染中行内代码和代码块出现黑色背景的问题——`NEUTRAL_MARKDOWN_THEME` 现在将所有 Rich 默认的 `markdown.*` 样式覆盖为 `"none"`，防止 Rich 内置的 `"cyan on black"` 在非黑色背景终端上泄露

src/kimi_cli/ui/shell/prompt.py

@@ -612,82 +612,15 @@ def _render_selected_item_lines(
 
 
 class LocalFileMentionCompleter(Completer):
-    """Offer fuzzy `@` path completion by indexing workspace files."""
+    """Offer fuzzy `@` path completion by indexing workspace files.
+
+    File discovery and ignore rules are delegated to
+    :mod:`kimi_cli.utils.file_filter` so that the web backend can reuse
+    them.
+    """
 
     _FRAGMENT_PATTERN = re.compile(r"[^\s@]+")
     _TRIGGER_GUARDS = frozenset((".", "-", "_", "`", "'", '"', ":", "@", "#", "~"))
-    _IGNORED_NAME_GROUPS: dict[str, tuple[str, ...]] = {
-        "vcs_metadata": (".DS_Store", ".bzr", ".git", ".hg", ".svn"),
-        "tooling_caches": (
-            ".build",
-            ".cache",
-            ".coverage",
-            ".fleet",
-            ".gradle",
-            ".idea",
-            ".ipynb_checkpoints",
-            ".pnpm-store",
-            ".pytest_cache",
-            ".pub-cache",
-            ".ruff_cache",
-            ".swiftpm",
-            ".tox",
-            ".venv",
-            ".vs",
-            ".vscode",
-            ".yarn",
-            ".yarn-cache",
-        ),
-        "js_frontend": (
-            ".next",
-            ".nuxt",
-            ".parcel-cache",
-            ".svelte-kit",
-            ".turbo",
-            ".vercel",
-            "node_modules",
-        ),
-        "python_packaging": (
-            "__pycache__",
-            "build",
-            "coverage",
-            "dist",
-            "htmlcov",
-            "pip-wheel-metadata",
-            "venv",
-        ),
-        "java_jvm": (".mvn", "out", "target"),
-        "dotnet_native": ("bin", "cmake-build-debug", "cmake-build-release", "obj"),
-        "bazel_buck": ("bazel-bin", "bazel-out", "bazel-testlogs", "buck-out"),
-        "misc_artifacts": (
-            ".dart_tool",
-            ".serverless",
-            ".stack-work",
-            ".terraform",
-            ".terragrunt-cache",
-            "DerivedData",
-            "Pods",
-            "deps",
-            "tmp",
-            "vendor",
-        ),
-    }
-    _IGNORED_NAMES = frozenset(name for group in _IGNORED_NAME_GROUPS.values() for name in group)
-    _IGNORED_PATTERN_PARTS: tuple[str, ...] = (
-        r".*_cache$",
-        r".*-cache$",
-        r".*\.egg-info$",
-        r".*\.dist-info$",
-        r".*\.py[co]$",
-        r".*\.class$",
-        r".*\.sw[po]$",
-        r".*~$",
-        r".*\.(?:tmp|bak)$",
-    )
-    _IGNORED_PATTERNS = re.compile(
-        "|".join(f"(?:{part})" for part in _IGNORED_PATTERN_PARTS),
-        re.IGNORECASE,
-    )
 
     def __init__(
         self,
@@ -701,9 +634,12 @@ def __init__(
         self._limit = limit
         self._cache_time: float = 0.0
         self._cached_paths: list[str] = []
+        self._cache_scope: str | None = None
         self._top_cache_time: float = 0.0
         self._top_cached_paths: list[str] = []
         self._fragment_hint: str | None = None
+        self._is_git: bool | None = None  # lazily detected
+        self._git_index_mtime: float | None = None
 
         self._word_completer = WordCompleter(
             self._get_paths,
@@ -717,21 +653,15 @@ def __init__(
             pattern=r"^[^\s@]*",
         )
 
-    @classmethod
-    def _is_ignored(cls, name: str) -> bool:
-        if not name:
-            return True
-        if name in cls._IGNORED_NAMES:
-            return True
-        return bool(cls._IGNORED_PATTERNS.fullmatch(name))
-
     def _get_paths(self) -> list[str]:
         fragment = self._fragment_hint or ""
         if "/" not in fragment and len(fragment) < 3:
             return self._get_top_level_paths()
         return self._get_deep_paths()
 
     def _get_top_level_paths(self) -> list[str]:
+        from kimi_cli.utils.file_filter import is_ignored
+
         now = time.monotonic()
         if now - self._top_cache_time <= self._refresh_interval:
             return self._top_cached_paths
@@ -740,7 +670,7 @@ def _get_top_level_paths(self) -> list[str]:
         try:
             for entry in sorted(self._root.iterdir(), key=lambda p: p.name):
                 name = entry.name
-                if self._is_ignored(name):
+                if is_ignored(name):
                     continue
                 entries.append(f"{name}/" if entry.is_dir() else name)
                 if len(entries) >= self._limit:
@@ -753,45 +683,45 @@ def _get_top_level_paths(self) -> list[str]:
         return self._top_cached_paths
 
     def _get_deep_paths(self) -> list[str]:
-        now = time.monotonic()
-        if now - self._cache_time <= self._refresh_interval:
-            return self._cached_paths
-
-        paths: list[str] = []
-        try:
-            for current_root, dirs, files in os.walk(self._root):
-                relative_root = Path(current_root).relative_to(self._root)
+        from kimi_cli.utils.file_filter import (
+            detect_git,
+            git_index_mtime,
+            list_files_git,
+            list_files_walk,
+        )
 
-                # Prevent descending into ignored directories.
-                dirs[:] = sorted(d for d in dirs if not self._is_ignored(d))
+        fragment = self._fragment_hint or ""
 
-                if relative_root.parts and any(
-                    self._is_ignored(part) for part in relative_root.parts
-                ):
-                    dirs[:] = []
-                    continue
+        scope: str | None = None
+        if "/" in fragment:
+            scope = fragment.rsplit("/", 1)[0]
 
-                if relative_root.parts:
-                    paths.append(relative_root.as_posix() + "/")
-                    if len(paths) >= self._limit:
-                        break
+        now = time.monotonic()
+        cache_valid = (
+            now - self._cache_time <= self._refresh_interval and self._cache_scope == scope
+        )
 
-                for file_name in sorted(files):
-                    if self._is_ignored(file_name):
-                        continue
-                    relative = (relative_root / file_name).as_posix()
-                    if not relative:
-                        continue
-                    paths.append(relative)
-                    if len(paths) >= self._limit:
-                        break
+        # Invalidate on .git/index mtime change (like Claude Code).
+        if cache_valid and self._is_git:
+            mtime = git_index_mtime(self._root)
+            if mtime != self._git_index_mtime:
+                cache_valid = False
 
-                if len(paths) >= self._limit:
-                    break
-        except OSError:
+        if cache_valid:
             return self._cached_paths
 
+        if self._is_git is None:
+            self._is_git = detect_git(self._root)
+
+        paths: list[str] | None = None
+        if self._is_git:
+            paths = list_files_git(self._root, scope)
+            self._git_index_mtime = git_index_mtime(self._root)
+        if paths is None:
+            paths = list_files_walk(self._root, scope, limit=self._limit)
+
         self._cached_paths = paths
+        self._cache_scope = scope
         self._cache_time = now
         return self._cached_paths