A Model Context Protocol server that provides filesystem operations including reading, writing, searching, and managing files and directories.
Server provides 10 filesystem tools with generally solid descriptions and basic schemas. Tools follow verb_noun naming convention (read_file, write_file, edit_file, etc.), which is good. However, several critical gaps reduce overall quality: (1) Input schemas are inferred from code snippets rather than explicitly visible in a structured tool registration block, limiting confidence in schema completeness. (2) Parameter descriptions exist but are minimal (10-30 chars), below the 72-char production baseline. (3) No output schemas documented, LLMs cannot infer what structure read_file or list_directory returns. (4) Error handling is generic ('Error: ' + message) with no recovery guidance or classification. (5) No tool annotations (readOnlyHint, destructiveHint, idempotentHint) despite clear risk profiles (READ_ONLY vs WRITE vs REVERSIBLE). (6) No dry-run confirmation pattern for destructive ops (write_file, delete via move). Tool descriptions are moderately comprehensive (100-200 chars) and state WHAT the tool does and WHEN to use it, which is positive. File operations are properly risk-classified in metadata but not exposed in schema annotations.
Create a new directory or ensure a directory exists. Can create multiple nested directories in one operation. If the directory already exists, this operation will succeed silently. Perfect for setting up directory structures for projects or ensuring required paths exist. Only works within allowed directories.
Get a recursive tree view of files and directories as a JSON structure. Each entry includes 'name', 'type' (file/directory), and 'children' for directories. Files have no children array, while directories always have a children array (which may be empty). The output is formatted with 2-space indentation for readability. Only works within allowed directories.
Make line-based edits to a text file. Each edit replaces exact line sequences with new content. Returns a git-style diff showing the changes made. Only works within allowed directories.
Retrieve detailed metadata about a file or directory. Returns comprehensive information including size, creation time, last modified time, permissions, and type. This tool is perfect for understanding file characteristics without reading the actual content. Only works within allowed directories.
Get a detailed listing of all files and directories in a specified path. Results clearly distinguish between files and directories with [FILE] and [DIR] prefixes. This tool is essential for understanding directory structure and finding specific files within a directory. Only works within allowed directories.
Output schemas not documented. LLMs cannot predict response structure for any tool. read_file returns string content, but list_directory returns formatted [FILE]/[DIR] text, directory_tree returns JSON. These differences are invisible in the API contract.
No tool annotations despite clear risk profiles. write_file, edit_file, move_file, create_directory should be marked with destructiveHint=true. read_file, read_multiple_files, list_directory, directory_tree, search_files, get_file_info should be marked with readOnlyHint=true. This prevents LLMs from accidentally triggering state-changing operations.
Inferred effective spec: <=2025-11-25.
| Scored | Grade | Overall | Spec posture | Rubric |
|---|---|---|---|---|
| 2026-09-22 | F | 49 | <=2025-11-25 | v2 |
| 2026-03-09 | D | 58 | - | v1 |
Move or rename files and directories. Can move files between directories and rename them in a single operation. If the destination exists, the operation will fail. Works across different directories and can be used for simple renaming within the same directory. Both source and destination must be within allowed directories.
Read the complete contents of a file from the file system. Handles various text encodings and provides detailed error messages if the file cannot be read. Use this tool when you need to examine the contents of a single file. Only works within allowed directories.
Read the contents of multiple files simultaneously. This is more efficient than reading files one by one when you need to analyze or compare multiple files. Each file's content is returned with its path as a reference. Failed reads for individual files won't stop the entire operation. Only works within allowed directories.
Recursively search for files and directories matching a glob pattern. Searches through all subdirectories from the starting path. Supports glob patterns like '*.java' or '*.{java,xml}'. Returns full paths to all matching items. Great for finding files when you don't know their exact location. Only searches within allowed directories.
Create a new file or completely overwrite an existing file with new content. Use with caution as it will overwrite existing files without warning. Handles text content with proper encoding. Only works within allowed directories.
No dry-run or confirmation pattern for destructive operations. write_file and edit_file support dryRun parameter (visible in edit_file schema), but write_file does not. No pre-execution confirmation for move_file (which can lose data if destination exists). Agents cannot safely preview changes before commit.
Error handling is generic. Code shows catch blocks returning 'Error: ' + e.getMessage() with isError=true. LLMs receive no guidance on whether errors are retryable, user-fixable, or fatal. No recovery hints ('Try search_files()' if file not found). No invalid value reporting.
Parameter descriptions are minimal. 'The file path to read' (24 chars) is below the 72-char production baseline. Missing format details: is path absolute or relative? Does it support ~ expansion? What encodings are supported for read_file? Are symlinks followed? These must be explicit to prevent invalid LLM calls.
edit_file parameter 'oldText' and 'newText' lack example patterns. LLMs will guess at multi-line matching behavior, regex vs literal, whitespace sensitivity. Description should specify: 'Exact string match (case-sensitive, whitespace-sensitive, can span multiple lines)' to prevent silent mismatches.
search_files glob pattern documentation is incomplete. Does it support ** (globstar)? Which glob syntax, Java nio.file.FileSystems or standard Unix? Excludepatterns are optional but no example given. LLMs will guess at syntax and fail silently.
read_multiple_files returns results but no output schema specified. Does it return one large blob? One struct per file? Does a single file failure fail the entire call or return partial results? Code shows failures are handled per-file ('Failed reads for individual files won't stop the entire operation'), but response format is invisible.
get_file_info returns 'comprehensive information including size, creation time, last modified time, permissions, and type' but response schema is undocumented. LLMs cannot extract specific fields or compose downstream calls. Is this JSON? Plaintext? What field names are used?