Name: terminal
Rating: 70.39999999999999 (1 reviews)
Author: wonderwhy-er

terminal

Use Desktop Commander for terminal and command-line work, especially anything that needs a shell whose state persists across turns: Python/Node REPLs, database shells, dev servers and other long-running processes, SSH into remote machines, and Windows PowerShell. Also handles everyday terminal tasks — navigating folders, choosing the right command for the user's shell (PowerShell, cmd, bash, zsh), running Docker/curl/cloud-CLI commands, inspecting processes and ports, and saving recurring workflows as scripts — even when the user doesn't say "Desktop Commander" or "terminal." Reach for it on pasted errors like "command not found", "permission denied", "EADDRINUSE", "address already in use", "npm ERR!", "ModuleNotFoundError", or "ENOENT", and on intents like "what's using port 3000", "kill that process", "ssh into my server", or "why won't my dev server start". Works on Windows, macOS, and Linux. Never run destructive commands without explicit confirmation.

Quality

86%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Passed

No findings from the security scan

The canonical home for this skill is terminal in wonderwhy-er/DesktopCommanderMCP

Quality

Content

81%Weight 40%Scale 1-5

Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.

The body is highly actionable — exact Desktop Commander call patterns, per-OS command tables, and a safety section with genuine validation checkpoints for destructive operations. Its weaknesses are efficiency (re-teaching Claude cross-shell equivalents and chaining semantics it already knows) and structure (a single 185-line file where a reference bundle would allow a leaner overview).

Suggestions

Delete or drastically shrink §2's equivalents table and the 'Notes that bite people' paragraph — Claude already knows PowerShell vs bash syntax; keep only the non-obvious rule (prefer cross-platform tools like python/node/git over shell-specific syntax) and the pointer to detect the shell via get_config.

Replace §7's chaining explanation ('&& runs the next only if the previous succeeded; ; runs regardless; || runs only on failure') with the one non-obvious fact (PowerShell hosts before v7 lack &&/||) and the destructive-command chaining rule.

Move the §2 equivalents table and §4 per-CLI details into references/ files (e.g. references/shell-commands.md, references/clis.md), leaving SKILL.md a lean overview with one-level-deep, clearly signaled references.

Dimension	Reasoning	Score
Conciseness	The Desktop Commander specifics (get_config fields, start_process/interact_with_process patterns, blockedCommands) earn their tokens, but several sections re-explain what Claude already knows: the §2 bash/PowerShell/cmd equivalents table, 'Notes that bite people' about Windows backslashes, and §7's '&& runs the next only if the previous succeeded; ; runs regardless' chaining semantics.	3 / 5
Actionability	Fully executable throughout: exact tool invocations ('call get_config and read systemInfo', 'start_process("python3 -i")', 'interact_with_process(pid, "...")') and copy-paste-ready per-OS commands ('lsof -i :3000', 'Get-NetTCPConnection -LocalPort 3000', 'netstat -ano \| findstr :3000'), covering the common cases including the full REPL workhorse pattern.	5 / 5
Workflow Clarity	Clear sequencing (detect environment → pick shell-correct command → run via the right process pattern → diagnose from real output), with validation checkpoints around destructive operations ('show the process line before terminating it', 'ls the glob before rm it', prefer dry-run first) and an explicit validate→name-cause→fix→retry feedback loop in §6, satisfying the destructive-ops validation requirement.	5 / 5
Progressive Disclosure	No bundle files exist, and the body is well-sectioned with clear §-cross-references; but at ~185 lines it exceeds the simple-skill case, and reference-style material (the §2 equivalents table, §4 CLI catalog, §7 destructive list) is inlined where a references/ file would keep SKILL.md a leaner overview — good structure with minor organization gaps.	4 / 5
	Total	17 / 20 Passed

Description

91%Weight 40%Scale 1-5

Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.

A strong description: it explicitly states both what the skill does and when to reach for it, backed by unusually good verbatim error-message and intent triggers. Its main weakness is verbosity — roughly 180 words where the good examples run 20-40 — which pads specificity rather than adding precision.

Suggestions

Trim the description to ~80-100 words by collapsing the two long task lists into one; e.g. 'Run terminal commands via Desktop Commander — persistent REPLs (Python/Node), database shells, dev servers, SSH, plus everyday shell tasks across PowerShell, cmd, bash, and zsh.'

Cut redundant trigger strings that duplicate each other (e.g. keep 'EADDRINUSE' or 'address already in use', not both) to reduce padding without losing keyword coverage.

Narrow the 'everyday terminal tasks' clause to reduce overlap risk with a generic shell/Bash skill — e.g. lead with the Desktop Commander tool requirement so the description reads as tool-usage, not general terminal help.

Dimension	Reasoning	Score
Specificity	Lists multiple concrete actions ('navigating folders', 'running Docker/curl/cloud-CLI commands, inspecting processes and ports, and saving recurring workflows as scripts') with near-comprehensive coverage, but the ~180-word description is noticeably padded, and the rubric instructs not to reward verbosity — so it sits between the 'several specific actions' and 'comprehensive coverage' anchors, above the midpoint.	4 / 5
Completeness	Explicitly answers what ('Use Desktop Commander for terminal and command-line work... Also handles everyday terminal tasks...') and when ('Reach for it on pasted errors like... and on intents like...') with concrete trigger phrases, plus an explicit trigger-guide for unnamed invocations ('even when the user doesn't say "Desktop Commander" or "terminal"').	5 / 5
Trigger Term Quality	Comprehensive natural-language triggers including literal error strings users paste ('command not found', 'permission denied', 'EADDRINUSE', 'npm ERR!', 'ModuleNotFoundError', 'ENOENT'), their plain-English synonyms ('address already in use'), and verbatim intents ('what's using port 3000', 'kill that process', 'ssh into my server', 'why won't my dev server start').	5 / 5
Distinctiveness Conflict Risk	Anchored to a specific tool (Desktop Commander) with distinct triggers (persistent REPLs, SSH sessions, port/process inspection), but 'everyday terminal tasks — navigating folders, choosing the right command' is broad enough to overlap with generic shell/Bash skills; mostly distinct with minor overlap risk.	4 / 5
	Total	18 / 20 Passed

Validation

93%

Warnings & errors only

Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.

Validation — 15 / 16 Passed

Validation for skill structure

Criteria	Description	Result
frontmatter_unknown_keys	Unknown frontmatter key(s) found; consider removing or moving to metadata	Warning

	Total	15 / 16 Passed

Repository: wonderwhy-er/DesktopCommanderMCP
Path: plugins/claude/skills/terminal/SKILL.md
Commit: 7bad545

Reviewed: 4 days ago

Table of Contents

Discovery Implementation Validation

Is this your skill?

If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.