WorkspaceGuard
Server Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
| WORKSPACEGUARD_TOKEN | No | Bearer token for HTTP authentication. Required when --transport http is used. | |
| WORKSPACEGUARD_ALLOWED_ORIGINS | No | Comma-separated list of allowed origins for CORS. | |
| WORKSPACEGUARD_OAUTH_APPROVAL_CODE | No | Approval code for OAuth-dev mode. Required when --auth-mode oauth-dev is used. |
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": true
} |
Tools
Functions exposed to the LLM to take actions
| Name | Description |
|---|---|
| workspaceguard_infoB | Return WorkspaceGuard runtime configuration summary. |
| policy_describeA | Show the current local policy profile for this WorkspaceGuard server. |
| echoA | Development smoke-test tool. |
| task_startC | Start an in-memory task under an open workspace. |
| task_updateC | Update task status or append a note. |
| task_statusA | Return one task by id, or all known in-memory tasks. |
| snapshot_createC | Create a file manifest snapshot for an open workspace. |
| checkpoint_createA | Record a named checkpoint for an existing snapshot id. This creates metadata only; restore is not implemented and snapshot existence is not validated yet. |
| drift_checkB | Create a fresh snapshot and compare it with the previous drift_check snapshot for the workspace. The first call records the baseline and reports no drift. |
| verification_runC | Run a structured verification command inside an open workspace. |
| workspace_openA | Open an allowed local checkout workspace and return a workspaceId. |
| workspace_statusB | Return a known workspace by id, or list open workspaces when workspaceId is omitted. |
| shell_runA | Run a structured command array inside an open workspace. This is real local execution. |
| git_statusA | Run git status --porcelain=v1 inside an open workspace. |
| git_diffC | Run git diff --no-color inside an open workspace. |
| file_readB | Read a UTF-8 file inside an open workspace. |
| directory_listC | List a directory inside an open workspace. |
| search_textB | Search for a literal text pattern inside an open workspace. |
| file_writeB | Create or overwrite a UTF-8 file inside an open workspace. |
| file_editA | Replace an exact text block that appears exactly once in a workspace file. |
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 20 tools
Most tools have clearly distinct purposes, but shell_run and verification_run overlap as both execute commands, and their descriptions do not clearly differentiate use cases. This could cause agent misselection.
The naming pattern is mixed: most tools use noun_verb (e.g., task_start, snapshot_create), but file operations use verb_noun (e.g., file_read, file_edit), and search_text is verb_noun while drift_check is noun_verb. This inconsistency, while readable, breaks a predictable pattern.
With 20 tools covering workspace management, file operations, git, snapshots, tasks, and execution, the count is slightly over the ideal 3-15 range but still reasonable for the scope and complexity of the server.
Several significant gaps exist: no workspace_close, no file delete, no snapshot restore (explicitly noted as unimplemented), no task delete, and no policy management beyond describe. These will likely cause agent failures when attempting standard workflows.