mle-kit-mcp 1.0.2__tar.gz → 1.1.0__tar.gz

{mle_kit_mcp-1.0.2 → mle_kit_mcp-1.1.0}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: mle-kit-mcp
-Version: 1.0.2
+Version: 1.1.0
 Summary: MCP server that provides different tools for MLE
 Author-email: Ilya Gusev <phoenixilya@gmail.com>
 Project-URL: Homepage, https://github.com/IlyaGusev/mle_kit_mcp

{mle_kit_mcp-1.0.2 → mle_kit_mcp-1.1.0}/mle_kit_mcp/tools/__init__.py

@@ -8,6 +8,7 @@ from .llm_proxy import (
     llm_proxy_local,
     llm_proxy_remote,
 )
+from .file_system import glob, grep
 
 
 __all__ = [
@@ -17,4 +18,6 @@ __all__ = [
     "remote_download",
     "llm_proxy_local",
     "llm_proxy_remote",
+    "glob",
+    "grep",
 ]

{mle_kit_mcp-1.0.2 → mle_kit_mcp-1.1.0}/mle_kit_mcp/tools/bash.py

@@ -1,6 +1,7 @@
 import atexit
 import signal
 import shlex
+import os
 from typing import Optional, Any
 
 from docker import from_env as docker_from_env  # type: ignore
@@ -26,6 +27,8 @@ def get_docker_client() -> DockerClient:
 
 def create_container() -> Container:
     client = get_docker_client()
+    uid = os.getuid()
+    gid = os.getgid()
     container = client.containers.run(
         BASE_IMAGE,
         "tail -f /dev/null",
@@ -33,6 +36,11 @@ def create_container() -> Container:
         remove=True,
         tty=True,
         stdin_open=True,
+        user=f"{uid}:{gid}",
+        environment={
+            "HOME": DOCKER_WORKSPACE_DIR_PATH,
+            "XDG_CACHE_HOME": f"{DOCKER_WORKSPACE_DIR_PATH}/.cache",
+        },
         volumes={
             get_workspace_dir(): {
                 "bind": DOCKER_WORKSPACE_DIR_PATH,

mle_kit_mcp-1.1.0/mle_kit_mcp/tools/file_system.py

@@ -0,0 +1,155 @@
+from typing import List, Optional
+from pathlib import Path
+import subprocess
+
+from mle_kit_mcp.files import get_workspace_dir
+
+
+def glob(pattern: str, path: Optional[str] = None) -> List[str]:
+    """
+    - Fast file pattern matching tool that works with any codebase size
+    - Supports glob patterns like "**/*.js" or "src/**/*.ts"
+    - Returns matching file paths sorted by modification time
+    - Use this tool when you need to find files by name patterns
+
+    Args:
+        pattern: The glob pattern to match files against, required.
+        path: The directory to search in. If not specified, the current working directory will be used.
+            IMPORTANT: Omit this field to use the default directory. DO NOT enter "undefined" or "null".
+            Simply omit it for the default behavior.
+            Must be a valid directory path if provided.
+
+    Returns:
+        A list of matching file paths.
+    """
+    full_path: Path = get_workspace_dir()
+    if path is not None:
+        full_path = get_workspace_dir() / path
+    files = [p for p in full_path.glob(pattern)]
+    files.sort(key=lambda p: p.stat().st_mtime, reverse=True)
+    resolved_files = [str(p.resolve()) for p in files]
+    resolved_files = [f.replace(str(get_workspace_dir()) + "/", "") for f in resolved_files]
+    return resolved_files
+
+
+def grep(
+    pattern: str,
+    path: Optional[str] = None,
+    glob: Optional[str] = None,
+    output_mode: str = "files_with_matches",
+    before_context: Optional[int] = None,
+    after_context: Optional[int] = None,
+    center_context: Optional[int] = None,
+    insensitive: bool = False,
+    type: Optional[str] = None,
+    head_limit: Optional[int] = None,
+    multiline: bool = False,
+) -> str:
+    r"""
+    A powerful search tool built on ripgrep
+
+    Usage:
+    - ALWAYS use Grep for search tasks. NEVER invoke `grep` or `rg` as a Bash command.
+    - The Grep tool has been optimized for correct permissions and access.
+    - Supports full regex syntax (e.g., "log.*Error")
+    - Filter files with glob parameter (e.g., "*.js", "**/*.tsx") or type parameter (e.g., "js", "py", "rust")
+    - Output modes: "content" shows matching lines, "files_with_matches" shows only file paths (default), "count" shows match counts
+    - Pattern syntax: Uses ripgrep (not grep) - literal braces need escaping (use `interface\{\}` to find `interface{}` in Go code)
+    - Multiline matching: By default patterns match within single lines only. For cross-line patterns like `struct \{[\s\S]*?field`, use `multiline: true`
+
+    Args:
+        pattern: The regular expression pattern to search for in file contents
+        path: File or directory to search. Defaults to current working directory.
+        glob: Glob pattern to filter files (e.g. "*.js", "*.{ts,tsx}")
+        output_mode: The output mode to use. Possible values are "content", "files_with_matches", "count".
+            "content" shows matching lines (supports -A/-B/-C context, -n line numbers, head_limit)
+            "files_with_matches" shows file paths (supports head_limit)
+            "count" shows match counts (supports head_limit).
+            Defaults to "files_with_matches"
+        before_context: Number of lines to show before each match (rg -B). Requires output_mode: "content", ignored otherwise.
+        after_context: Number of lines to show after each match (rg -A). Requires output_mode: "content", ignored otherwise.
+        center_context: Number of lines to show before and after each match (rg -C). Requires output_mode: "content", ignored otherwise.
+        insensitive: Case insensitive search (rg -i). Defaults to False.
+        type: File type to search (rg --type). Common types: js, py, rust, go, java, etc. More efficient than include for standard file types.
+        head_limit: Limit output to first N lines/entries, equivalent to "| head -N". Works across all output modes: content (limits output lines), files_with_matches (limits file paths), count (limits count entries). When unspecified, shows all results from ripgrep.
+        multiline: Enable multiline mode where . matches newlines and patterns can span lines (rg -U --multiline-dotall). Default: false.
+
+    Returns:
+        A string with the search results.
+    """
+    assert pattern, "'pattern' must be a non-empty string"
+
+    valid_output_modes = {"content", "files_with_matches", "count"}
+    assert (
+        output_mode in valid_output_modes
+    ), f"Invalid output_mode: {output_mode}. Expected one of {sorted(valid_output_modes)}"
+
+    if center_context is not None:
+        assert (
+            before_context is None and after_context is None
+        ), "Use either 'center_context' or ('before_context'/'after_context'), not both"
+
+    cmd: List[str] = [
+        "rg",
+        "--color=never",
+        "--no-heading",
+    ]
+
+    if output_mode == "files_with_matches":
+        cmd.append("-l")
+    elif output_mode == "count":
+        cmd.append("-c")
+    else:
+        cmd.append("-n")
+
+    if insensitive:
+        cmd.append("-i")
+
+    if multiline:
+        cmd.extend(["--multiline", "--multiline-dotall"])
+
+    if type:
+        cmd.extend(["--type", type])
+
+    if glob:
+        cmd.extend(["--glob", glob])
+
+    if output_mode == "content":
+        if center_context is not None:
+            cmd.extend(["-C", str(center_context)])
+        else:
+            if before_context is not None:
+                cmd.extend(["-B", str(before_context)])
+            if after_context is not None:
+                cmd.extend(["-A", str(after_context)])
+
+    cmd.append("--")
+    cmd.append(pattern)
+
+    search_root: Path = get_workspace_dir()
+    if path is not None:
+        search_root = search_root / path
+    cmd.append(str(search_root))
+
+    try:
+        result = subprocess.run(
+            cmd,
+            stdout=subprocess.PIPE,
+            stderr=subprocess.PIPE,
+            text=True,
+            check=False,
+        )
+    except FileNotFoundError:
+        return "ripgrep (rg) not found. Please install ripgrep in the environment where this tool runs."
+
+    stdout = (result.stdout or "").rstrip("\n")
+    stderr = (result.stderr or "").strip()
+
+    if head_limit is not None and head_limit >= 0 and stdout:
+        stdout_lines = stdout.splitlines()
+        stdout = "\n".join(stdout_lines[:head_limit])
+
+    if result.returncode not in (0, 1) and stderr:
+        return stderr
+
+    return stdout

{mle_kit_mcp-1.0.2 → mle_kit_mcp-1.1.0}/mle_kit_mcp.egg-info/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: mle-kit-mcp
-Version: 1.0.2
+Version: 1.1.0
 Summary: MCP server that provides different tools for MLE
 Author-email: Ilya Gusev <phoenixilya@gmail.com>
 Project-URL: Homepage, https://github.com/IlyaGusev/mle_kit_mcp

{mle_kit_mcp-1.0.2 → mle_kit_mcp-1.1.0}/mle_kit_mcp.egg-info/SOURCES.txt

@@ -16,10 +16,12 @@ mle_kit_mcp.egg-info/requires.txt
 mle_kit_mcp.egg-info/top_level.txt
 mle_kit_mcp/tools/__init__.py
 mle_kit_mcp/tools/bash.py
+mle_kit_mcp/tools/file_system.py
 mle_kit_mcp/tools/llm_proxy.py
 mle_kit_mcp/tools/remote_gpu.py
 mle_kit_mcp/tools/text_editor.py
 tests/test_bash.py
+tests/test_file_system.py
 tests/test_llm_proxy.py
 tests/test_text_editor.py
 tests/test_truncate_context.py

{mle_kit_mcp-1.0.2 → mle_kit_mcp-1.1.0}/pyproject.toml

@@ -4,7 +4,7 @@ build-backend = "setuptools.build_meta"
 
 [project]
 name = "mle-kit-mcp"
-version = "1.0.2"
+version = "1.1.0"
 description = "MCP server that provides different tools for MLE"
 readme = "README.md"
 authors = [

{mle_kit_mcp-1.0.2 → mle_kit_mcp-1.1.0}/tests/test_bash.py

@@ -33,3 +33,10 @@ def test_bash_timeout_with_output() -> None:
     result = bash("echo 'hello' && sleep 100", timeout=5)
     assert "hello" in result
     assert "Command timed out" in result
+
+
+def test_bash_ownership() -> None:
+    bash("touch dummy")
+    assert os.path.exists(get_workspace_dir() / "dummy")
+    assert os.stat(get_workspace_dir() / "dummy").st_uid == os.getuid()
+    assert os.stat(get_workspace_dir() / "dummy").st_gid == os.getgid()

mle_kit_mcp-1.1.0/tests/test_file_system.py

@@ -0,0 +1,74 @@
+
+import shutil
+import pytest
+
+from mle_kit_mcp.tools import glob, bash, grep
+
+rg_missing = shutil.which("rg") is None
+
+
+def test_glob_base():
+    bash("mkdir -p dummy_dir")
+    bash("touch dummy.txt", cwd="dummy_dir")
+    assert glob("**/*.txt") == ["dummy_dir/dummy.txt"]
+
+
+@pytest.mark.skipif(rg_missing, reason="ripgrep (rg) is not installed")
+def test_grep_files_with_matches_basic():
+    bash("mkdir -p rg_dummy")
+    bash("printf 'hello\\nworld\\n' > a.txt", cwd="rg_dummy")
+    bash("printf 'print(\"Hello\")\\n' > b.py", cwd="rg_dummy")
+    bash("printf 'bye\\n' > c.txt", cwd="rg_dummy")
+
+    out = grep("hello", path="rg_dummy")
+    assert "a.txt" in out
+    assert "b.py" not in out
+
+
+@pytest.mark.skipif(rg_missing, reason="ripgrep (rg) is not installed")
+def test_grep_content_mode_with_context_and_insensitive():
+    bash("mkdir -p rg_dummy2")
+    bash("printf 'alpha\\nBravo\\ncharlie\\n' > notes.txt", cwd="rg_dummy2")
+    bash("printf 'print(\"Hello\")\\n' > code.py", cwd="rg_dummy2")
+
+    out = grep(
+        "hello",
+        path="rg_dummy2",
+        output_mode="content",
+        center_context=1,
+        insensitive=True,
+    )
+    assert "code.py" in out
+    assert ":" in out
+
+
+@pytest.mark.skipif(rg_missing, reason="ripgrep (rg) is not installed")
+def test_grep_count_mode_with_glob_and_head_limit():
+    bash("mkdir -p rg_dummy3")
+    bash("printf 'hello\\nworld\\n' > a.txt", cwd="rg_dummy3")
+    bash("printf 'foo\\nbar\\n' > m.txt", cwd="rg_dummy3")
+
+    out = grep(
+        "o",
+        path="rg_dummy3",
+        output_mode="count",
+        glob="*.txt",
+        head_limit=1,
+    )
+    lines = [line for line in out.splitlines() if line.strip()]
+    assert len(lines) <= 1
+    assert ".txt:" in lines[0]
+
+
+@pytest.mark.skipif(rg_missing, reason="ripgrep (rg) is not installed")
+def test_grep_multiline_matching():
+    bash("mkdir -p rg_dummy4")
+    bash("printf 'foo\\nbar\\n' > multi.txt", cwd="rg_dummy4")
+
+    out = grep(
+        "foo.*bar",
+        path="rg_dummy4",
+        output_mode="content",
+        multiline=True,
+    )
+    assert "multi.txt" in out