MCP Simple Shell
Provides tools for interacting with SQLite databases: creating database files, opening existing databases (optionally read-only), executing SQL statements, inspecting the full schema catalog, retrieving table column information, and closing connections.
This server does not currently have a deployable release on Glama. Follow the repository's setup instructions to install and configure it for your MCP client, including any required credentials and local dependencies.
MCP Simple Shell
A workspace-confined file operations server for the Model Context Protocol (MCP). Provides safe, sandboxed file and directory operations with built-in safety features.
Features
File Operations: Read, write, copy, move, delete, append, rename, touch, stat
Directory Operations: Create, delete, copy, move, rename, list
Text Processing: Replace sections, apply unified diff patches
Binary Operations: Read/write binary, hexdump, find byte patterns, diff binaries
Advanced Processing: UPX unpacking, XOR/AES decryption, YARA scanning, file type detection, firmware carving, entropy analysis, ELF/PE segment splitting
SQLite Database: Create, open, query, schema inspection
System Tools: Cryptographic hashing, PE rebasing, disk usage, server info
Trash System: Protected trash with SHA256 verification for recovery
Advisory Locking: Shared/exclusive file locking for coordination
Chunked Upload: Handle files larger than 64KB tool call limit
Related MCP server: Local Files MCP Server
Safety Features
No permanent deletion: All deletions go to protected trash with SHA256 verification
Atomic writes: Temp file + os.replace() for crash-safe writes
Symlink protection: O_NOFOLLOW + lstat() on all path components
Path containment: Workspace-root confinement, no traversal
Rate limiting: 100 ops/sec per session
Size limits: 10MB max file size, 8KB chunk limit for tool calls
Installation
# Install Python dependencies
pip install -r requirements.txt
# Install system dependencies (Ubuntu/Debian)
sudo scripts/install_deps.shUsage
# Run the MCP server
python -m mcp_simple_shell.server
# Or use the entry point
mcp-simple-shellThe server runs on streamable-http transport by default.
Tools
File Operations
file_read- Read text files (with encoding support)file_read_range- Read specific byte rangesfile_write- Create or overwrite filesfile_overwrite- Replace existing files (trash old version)file_append- Append to filesfile_rename- Rename filesfile_delete- Move to trashfile_copy- Copy filesfile_move- Move/rename filesfile_diff- Unified diff between filesfile_touch- Create empty file or update mtimefile_stat- Get file metadata
Directory Operations
make_dir- Create directoriesdelete_dir- Delete directories (with recursive trash)copy_dir- Copy directory treesmove_dir- Move directory treesrename_dir- Rename directorieslist_dir- List directory contentsfile_list- Recursive listing with filters
Text Processing
file_replace_section- Exact string replacementfile_patch- Apply unified diff patches
Binary Operations
file_read_binary- Read raw bytes (hex + ASCII)file_write_binary- Write raw bytes at offsetfile_hexdump- Formatted hex dumpfile_find_bytes- Pattern search with wildcardsfile_diff_binary- Byte-by-byte comparison
Advanced Processing
file_unpack- UPX unpackingfile_decrypt- XOR/AES-CBC decryptionfile_extract_range- Extract byte rangesfile_yara_scan- YARA rule scanningfile_magic- File type detection (libmagic)file_carve_firmware- Firmware extraction (binwalk)file_find_embedded- Entropy-based embedded file detectionfile_split_segments- ELF/PE segment splitting
SQLite
sqlite_create_file- Create databasesqlite_open_file- Open database (read-only option)sqlite_get_catalog- Full schema catalogsqlite_execute- Execute SQLsqlite_close- Close connectionsqlite_table_info- Table column info
System
file_hash- SHA256/MD5/SHA1/SSDEEP/TLSHfile_rebase- PE image base changedisk_usage- Workspace statisticsserver_info- Server configuration
Trash
list_trash- List trashed filestrash_info- View trash metadatarestore_from_trash- Restore with SHA256 verification
Locking
file_lock- Acquire advisory lockfile_unlock- Release advisory lock
Chunked Upload
begin_upload- Start upload sessionappend_chunk- Append chunkfinish_upload- Finalize uploadcancel_upload- Cancel upload
Configuration
Workspace root: ./workspace (created automatically)
Trash directory: ./workspace/.trash (managed only by trash tools)
Requirements
Python 3.10+
Linux (primary), macOS (limited binary tool support)
System packages: upx, binwalk, yara, libyara-dev, file, python3-dev, build-essential, libfuzzy-dev
License
MIT
Tool Schema Changelog
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
Loading tool schema changelog…
This server cannot be deployed
Maintenance
Related MCP Connectors
Browse and manage files in your Moxt AI workspace from any MCP client.
Manage files and folders directly from your workspace. Read and write files, list directories, cre…
Persistent file storage for AI agents via MCP and curl. Upload, download, and version files.
Read and write shared BitsWeave context, projects, tasks, and work sessions through MCP.
Related MCP Servers
- FlicenseNot gradedqualityDmaintenanceEnables file system operations such as listing, reading, and creating files within a scoped local project directory. It provides a secure way to manage local files through standardized MCP tools built with FastMCP.-
- AlicenseAqualityDmaintenanceProvides safe local file operations through MCP, including reading, writing, searching, organizing, and protected deletion with configurable path restrictions.121ISC
- FlicenseAqualityCmaintenanceEnables saving and managing files in a sandboxed local folder via MCP, with tools for listing, reading, writing, appending, searching, moving, and deleting files and directories.10-
- AlicenseNot gradedqualityDmaintenanceEnables ChatGPT or any MCP client to operate safely on a designated workspace by listing, reading, searching, writing, and trashing files, inspecting Git status/log/diff, and optionally running allowlisted executables without a shell.Apache 2.0