cmux

Control the cmux macOS terminal app: workspaces, panes, surfaces, browsers, notifications, settings, and hooks. Use only when the user explicitly names cmux. Read before running any cmux command.

Quality

96%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Low

Low-risk findings worth noting

Quality

Content

100%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 skill body is exemplary: dense, fully executable command guidance with precise safety rules (workspace anchoring, ownership, focus preservation), explicit re-read/re-list verification loops, and a clean progressive-disclosure split into two verified reference files. No meaningful weaknesses were found against any dimension.

Dimension	Reasoning	Score
Conciseness	Every line is operational (targeting rules, exact commands, flag constraints like "read-screen and capture-pane accept --workspace or --surface, not --pane"); there is no conceptual padding Claude already knows and no trimming opportunities, matching the lean anchor.	5 / 5
Actionability	Fully executable, copy-paste-ready commands throughout (e.g. `cmux read-screen --surface surface:7 --lines 100`, `cmux new-pane --workspace "$CMUX_WORKSPACE_ID" --type terminal --direction right --focus false`), with explicit gotchas naming nonexistent commands ("send-surface and send-key-surface do not exist") covering the common cases.	5 / 5
Workflow Clarity	The inspect/read/send flow is clearly sequenced (ping, identify, tree, list, read, send) and carries explicit validation checkpoints and feedback loops: "Re-read terminal output after sending input to check the result", "check cmux surface-health before sending input when state may be stale", and "Re-list panes and surfaces after layout changes", plus a troubleshooting section — matching the validate-act-verify anchor.	5 / 5
Progressive Disclosure	The body is a clear overview that splits detail into two real, one-level-deep references, each well-signaled per task ("Read the relevant section before starting these tasks"); all cited anchors (references/advanced.md#layout-and-sidebar-commands, #configuration, #socket-api-and-access-modes, references/viewers.md#browser-workflow, #markdown-workflow) resolve to existing headings.	5 / 5
	Total	20 / 20 Passed

Description

92%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.

The description is a strong example: it states what the skill controls with a comprehensive enumeration of capability areas, and gives an explicit, restrictive when-clause tied to the product name. The only minor gap is that natural-language coverage beyond the name "cmux" itself is thin, though that is partly inherent to a named-app skill.

Dimension	Reasoning	Score
Specificity	"Control the cmux macOS terminal app: workspaces, panes, surfaces, browsers, notifications, settings, and hooks" enumerates the full capability surface of the app with no apparent gaps, matching the comprehensive-coverage anchor rather than the level-4 anchor which requires minor gaps.	5 / 5
Completeness	Both parts are explicit and concrete: what ("Control the cmux macOS terminal app: workspaces, panes, surfaces, browsers, notifications, settings, and hooks") and when ("Use only when the user explicitly names cmux"), matching the anchor-5 pattern of clear what-plus-when with concrete trigger phrases.	5 / 5
Trigger Term Quality	"cmux" appears as an explicit, repeated natural trigger ("Use only when the user explicitly names cmux", "any cmux command") alongside "macOS terminal app", giving good keyword coverage; a few natural phrasings beyond the product name (e.g. terminal multiplexer, splitting the terminal) are absent, so it falls just short of the comprehensive-synonyms anchor.	4 / 5
Distinctiveness Conflict Risk	The skill targets a named proprietary app and explicitly scopes triggering to "Use only when the user explicitly names cmux", giving it a clear niche with minimal risk of firing for the wrong skill.	5 / 5
	Total	19 / 20 Passed

Validation

100%

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 — 16 / 16 Passed

Validation for skill structure

No warnings or errors.

Repository: davidondrej/skills
Path: skills/agent-orchestration/cmux/SKILL.md
Commit: f025cb4

Reviewed: 7 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.