Local Workspace MCP
Server Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
No arguments | |||
Instructions
Guidance the server publishes about itself, which clients place ahead of the tool catalog so the model reads it before choosing anything.
This server publishes no instructions, or was last inspected before Glama recorded them.
Capabilities
Features and capabilities supported by this server
Protocol revision2025-11-25
| Capability | Details |
|---|---|
| tools | {
"listChanged": false
} |
| prompts | {
"listChanged": false
} |
| resources | {
"subscribe": false,
"listChanged": false
} |
| experimental | {} |
Tools
Functions exposed to the LLM to take actions
| Name | Description |
|---|---|
| list_paired_devicesB | List local and owner-configured devices by name; never disclose SSH credentials. |
| current_user_infoB | Local OS identity. This project has no hosted subscription or vendor account. |
| ping_deviceC | Check local availability or establish an authenticated configured SSH MCP session. |
| list_device_toolsB | Read the actual tool schemas of a named paired device. |
| call_device_toolA | Execute a tool on a named paired device. Inspect list_device_tools first. This can change files, run commands, or stop processes with the remote user's permissions. |
| shutdown_device_agentA | Stop ONLY this MCP agent process. It does not shut down the computer. An SSH-launched on-demand agent may start again on the next connection. |
| list_directoryA | List supported text files and folders in a relative workspace path. Hidden files are excluded. |
| read_textA | Read a UTF-8 text file, max 1 MiB. Paths are relative to the configured workspace. |
| get_workflow_instructionsB | Get guidance for document generation, data analysis, verification, and delivery. |
| get_artifact_pathA | Return a verified artifact path for local clients; remote clients should use download links. |
| list_artifactsA | List Office documents, PDFs, images and text deliverables directly inside exports. |
Prompts
Interactive templates invoked by user choice
| Name | Description |
|---|---|
No prompts | |
Resources
Contextual data attached and managed by the client
| Name | Description |
|---|---|
No resources | |
TDQS
Scored across 11 tools
Each tool targets a distinct resource/action: identity, device listing/pinging, device tool discovery/execution, agent shutdown, local file listing/reading, workflow guidance, and artifact listing/path. Minor potential confusion among list-style tools and between local vs remote file operations is mostly resolved by descriptions.
Mostly consistent snake_case with verb_noun patterns such as list_*, get_*, read_text, call_device_tool, and shutdown_device_agent. current_user_info is a noun phrase rather than get_current_user_info, a minor deviation.
11 tools is well-scoped for a local workspace/device bridge: enough for device lifecycle, file access, and artifacts without excessive surface. No obvious redundant or bloat tools.
The set covers reading/listing local files and artifacts plus remote device discovery/execution, but lacks local write/create/update/delete operations and artifact creation/upload/download. These are notable gaps for a workspace server, though remote call_device_tool offers some execution coverage.