Skip to main content
Glama

Manage workspace files and notebooks

manage_workspace_files
Destructive

Manage Databricks workspace files, notebooks, and folders: list, inspect, export/import content, create directories, or delete objects via the Workspace API.

Instructions

Manage Databricks workspace files, notebooks and folders (Workspace API).

  • list (path[, recursive]), get_status (path): metadata (type, language, size, object_id).

  • export (path[, format, local_path]): text content inline (UTF-8) or base64 for binary; capped by DBX_MCP_MAX_INLINE_DOWNLOAD_BYTES; with local_path the file is written under DBX_MCP_LOCAL_FILE_ROOT.

  • import (path, content | content_base64 | local_path[, language, format, overwrite]): create or update a file or notebook (10 MB limit). Notebooks: language=PYTHON|SQL|SCALA|R with format SOURCE (default when language is set) or JUPYTER (.ipynb content). overwrite=true is DESTRUCTIVE (confirm required).

  • mkdirs (path): create directory and parents.

  • delete (path[, recursive]): DESTRUCTIVE, confirm required.

Safety classification: depends on input (DESTRUCTIVE, READ_ONLY, WRITE).

Input Schema

TableJSON Schema
NameRequiredDescriptionDefault
pathYesAbsolute workspace path, e.g. /Users/me@x.com/project/nb.
actionYeslist: directory contents; get_status: object metadata; export: download content (inline or to local_path); import: upload/create/update a file or notebook; mkdirs: create directories; delete: delete object (recursive for non-empty directories).
formatNoexport/import format. Export default: SOURCE for notebooks, AUTO otherwise. Import default: SOURCE when language is set (notebook), else AUTO (file, or notebook if the content has a notebook header). RAW is import-only.
confirmNoSet to true ONLY after the user has reviewed the plan returned by a previous call with status 'confirmation_required'. Required for destructive/security-sensitive actions.
contentNoimport: text content (UTF-8).
dry_runNoIf true, validate and return the planned change without executing it.
languageNoimport: notebook language (required for a single SOURCE notebook).
overwriteNoimport: replace an existing object (DESTRUCTIVE); export: replace a local file.
page_sizeNoMax items to return (server caps this).
recursiveNolist: walk sub-directories (files only); delete: delete non-empty directory.
local_pathNoimport: read from / export: write to this path relative to DBX_MCP_LOCAL_FILE_ROOT.
page_tokenNonext_page_token from a previous response.
content_base64Noimport: binary content, base64-encoded.
create_parentsNoimport: create missing parent directories.

Output Schema

TableJSON Schema
NameRequiredDescriptionDefault
dataNo
pageNo
planNo
toolYes
actionNo
safetyNo
statusNosuccess
summaryYes
warningsNo
next_stepsNoSuggested follow-up calls.
request_idNo

Schema Changelog

Changes observed during successful MCP inspections.

  1. First observedv0.1.0

TDQS

A4.7/5.0
Behavior5/5

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

Beyond the blanket destructiveHint=true annotation, the description pinpoints exactly which operations are destructive (import with overwrite, delete), states that overwrite/delete require confirmation, and discloses operational limits (DBX_MCP_MAX_INLINE_DOWNLOAD_BYTES, 10 MB import cap, DBX_MCP_LOCAL_FILE_ROOT scoping, inline UTF-8 vs base64 export). This is substantive behavioral context the annotations cannot express.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Conciseness5/5

Is the description appropriately sized, front-loaded, and free of redundancy?

One framing sentence followed by tight per-action bullets, front-loading the overall purpose and then the safety classification. For a six-action, 14-parameter tool the length is proportionate and every line carries action-specific information.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Completeness5/5

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

Given high complexity, an output schema (which removes the need to describe return values), and a confirmation workflow, the description covers all actions, destructive-warning semantics, defaults, and resource scoping. An agent has everything needed to select an action and invoke it safely.

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Parameters4/5

Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

With 100% schema coverage the baseline is 3, but the description adds genuine meaning by grouping parameters per action (which flags apply to export vs import vs list) and explaining the language/format interaction and defaults (SOURCE for single notebook, JUPYTER for .ipynb). It layers action-scoped semantics on top of the per-parameter schema text.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Purpose5/5

Does the description clearly state what the tool does and how it differs from similar tools?

The description names a specific resource (Databricks workspace files, notebooks, folders via the Workspace API) and enumerates each of the six actions with its distinct effect. An agent can distinguish workspace-file operations from the sibling manage_volume_files/manage_workspace without opening a schema.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Usage Guidelines4/5

Does the description explain when to use this tool, when not to, or what alternatives exist?

Each action line pairs the verb with its parameters and scope (e.g., 'list (path[, recursive])', 'delete (path[, recursive]): DESTRUCTIVE'), giving clear per-action context and explicitly flagging when overwrite/delete require confirmation. It stops short of naming alternatives or when-not-to-use conditions versus sibling tools, so it does not reach a 5.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.