io.github.crunchtools/systemd
Server Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
| SYSTEMD_UNIT_DIR | No | Directory unit_file_write/remove operate on | /etc/systemd/system |
| DBUS_SYSTEM_BUS_SOCKET | No | D-Bus system bus socket path | /run/dbus/system_bus_socket |
| SYSTEMD_EXTRA_PROTECTED_UNITS | No | Comma-separated units added to the built-in denylist |
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
} |
| logging | {} |
| prompts | {
"listChanged": false
} |
| resources | {
"subscribe": false,
"listChanged": false
} |
Tools
Functions exposed to the LLM to take actions
| Name | Description |
|---|---|
| unit_list_toolB | List systemd units. |
| unit_status_toolA | Get curated status for a unit: active/sub/load state, PID, memory, CPU. |
| unit_show_toolA | Dump the full property set for a unit (dependencies, exec settings, cgroup, etc). |
| unit_start_toolC | Start a unit. |
| unit_stop_toolB | Stop a unit. Refused for units on the protected list. |
| unit_restart_toolB | Restart a unit. Refused for units on the protected list. |
| unit_reload_toolA | Ask a unit to reload its configuration without restarting. |
| unit_enable_toolC | Enable a unit to start on boot. |
| unit_disable_toolB | Disable a unit from starting on boot. Refused for units on the protected list. |
| unit_mask_toolA | Mask a unit so it cannot be started even manually. Refused for protected units. |
| unit_unmask_toolA | Unmask a previously masked unit. |
| daemon_reload_toolA | Reload systemd's unit file cache (equivalent to systemctl daemon-reload). Returns: Reload confirmation |
| unit_file_write_toolA | Write a new unit file for setting up a service. Backs up any file it overwrites. Requires the host's unit directory (default /etc/systemd/system) to be bind-mounted into the container. Runs daemon-reload after writing. |
| unit_file_remove_toolA | Decommission a unit: stop, disable, optionally mask, back up and remove its file. Refused for units on the protected list. Best-effort on stop/disable — a unit that's already stopped or was never enabled doesn't block file removal. |
| journal_query_toolA | Query the journal — the main 2AM troubleshooting tool. |
| failed_units_toolA | List units currently in the 'failed' state — the first thing to check at 2AM. Returns: List of failed units with count |
| list_jobs_toolA | List pending systemd jobs — reveals stuck starts/stops/reloads. Returns: List of pending jobs with count |
| timer_list_toolA | List systemd timers with their next and last elapse times. Returns: List of timers with count |
| system_status_toolA | Get overall systemd manager state: running/degraded, failed and job counts. Returns: SystemState, Version, NFailedUnits, NJobs, NNames |
| hostinfo_toolA | Get host identity info: hostname, kernel, OS. Returns: Hostname, kernel, and OS details |
| session_list_toolA | List logged-in sessions — who's on this box right now. Returns: List of sessions with count |
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 21 tools
The unit lifecycle tools are each clearly scoped, and the file, journal, timer, and host/system tools query distinct areas. The only near-overlaps are unit_status_tool vs unit_show_tool and unit_reload_tool vs daemon_reload_tool, but their descriptions resolve the difference.
Most tools follow an object-first snake_case pattern with a _tool suffix, but non-unit tools break the consistency: unit_list_tool, list_jobs_tool, and timer_list_tool mix word order, while failed_units_tool and hostinfo_tool are noun phrases rather than actions. The names remain readable and predictable enough to navigate.
At 21 tools, this is above the typical 3-15 sweet spot, but nearly every tool covers a distinct lifecycle operation or status query for a systemd-focused server. The unit, unit-file, journal, and monitoring subgroups each add meaningful capability without obvious redundancy.
The unit surface covers the full lifecycle: write, start, stop, restart, reload, enable, disable, mask, unmask, show, and remove, plus daemon-reload. Monitoring is also well covered with journal queries, failed units, pending jobs, timers, system status, sessions, and host info, leaving no obvious dead ends for routine systemd administration.