tmux
Patterns for running long-lived processes in tmux. Use when starting dev servers, watchers, tilt, or any process expected to outlive the conversation.
88
83%
Does it follow best practices?
Impact
100%
1.14xAverage score across 3 eval scenarios
Passed
No known issues
Quality
Discovery
89%Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.
This is a solid description with clear 'Use when' guidance and good trigger terms that cover the main use cases. Its main weakness is the vague 'Patterns for' phrasing which doesn't specify concrete actions Claude would take (e.g., creating sessions, detaching, checking output). Overall it performs well for skill selection purposes.
Suggestions
Replace 'Patterns for running' with specific actions like 'Creates and manages tmux sessions for running long-lived processes' to improve specificity.
| Dimension | Reasoning | Score |
|---|---|---|
Specificity | Names the domain (long-lived processes in tmux) and gives examples (dev servers, watchers, tilt), but doesn't list specific concrete actions like 'create tmux sessions', 'detach processes', or 'monitor output'. The word 'Patterns' is vague. | 2 / 3 |
Completeness | Clearly answers both what ('Patterns for running long-lived processes in tmux') and when ('Use when starting dev servers, watchers, tilt, or any process expected to outlive the conversation') with an explicit 'Use when' clause. | 3 / 3 |
Trigger Term Quality | Includes strong natural trigger terms users would actually say: 'dev servers', 'watchers', 'tilt', 'tmux', 'long-lived processes', and 'outlive the conversation'. These cover common variations of how users would describe this need. | 3 / 3 |
Distinctiveness Conflict Risk | Highly distinctive niche: tmux + long-lived processes is a very specific combination unlikely to conflict with other skills. The trigger terms (dev servers, watchers, tilt, outlive the conversation) clearly delineate this skill's scope. | 3 / 3 |
Total | 11 / 12 Passed |
Implementation
77%Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.
This is a strong, highly actionable skill with excellent workflow clarity and concrete executable examples throughout. Its main weakness is verbosity from repeated boilerplate (the SESSION derivation line appears in nearly every code block) and some overlapping patterns that could be consolidated or split into a reference file. The hard requirements at the top and the decision table for when to use tmux are particularly effective.
Suggestions
Define the SESSION derivation once at the top and reference it in subsequent examples (e.g., 'Using SESSION as defined above') to reduce repetition across 12+ code blocks.
Consider extracting the 'Common Patterns' and 'Monitoring Output' sections into a separate TMUX_PATTERNS.md reference file, keeping SKILL.md as a concise overview with the core rules and one or two key examples.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The skill is mostly efficient with executable examples, but there's significant repetition — the SESSION derivation line appears 12+ times, and several patterns (single process, common patterns) overlap substantially. The 'when to use tmux' table and isolation rules are useful but the overall content could be tightened by ~30%. | 2 / 3 |
Actionability | Every section provides fully executable, copy-paste-ready bash commands with concrete examples. The WRONG vs CORRECT comparison, specific tmux flags, and real-world commands (npm run dev, tilt up, tail -f) make this highly actionable. | 3 / 3 |
Workflow Clarity | Multi-step processes are clearly sequenced with explicit validation checkpoints: check session exists before creating, check window exists before adding, poll for readiness before proceeding. The idempotent patterns with if/else guards and the restart sequence (Ctrl+C → sleep → restart) demonstrate proper feedback loops. | 3 / 3 |
Progressive Disclosure | The content is well-organized with clear section headers and a logical progression from rules → basics → monitoring → lifecycle → patterns. However, at ~150 lines of executable content, some sections (like the common patterns at the end) could be split into a separate reference file. Everything is inline in a single document. | 2 / 3 |
Total | 10 / 12 Passed |
Validation
100%Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.
Validation — 11 / 11 Passed
Validation for skill structure
No warnings or errors.
- Repository
- NeverSight/skills_feed
- Commit
e437c3c
- Reviewed
Table of Contents
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.