nw-ab-critique-dimensi…
Review dimensions for validating agent quality - template compliance, safety, testing, and priority validation
Terminal UI and CLI design patterns for product owners. Load when designing command-line tools, interactive terminal applications, or writing CLI-specific acceptance criteria.
$ npx -y skills add nWave-ai/nWave --skill nw-ux-tui-patterns --agent claude-codeHow it fires
How this skill gets triggered: by you, by Claude, or both.
/nw-ux-tui-patternsContext preview
The summary Claude sees to decide when to auto-load this skill.
Terminal UI and CLI design patterns for product owners. Load when designing command-line tools, interactive terminal applications, or writing CLI-specific acceptance criteria.
name: nw-ux-tui-patterns description: Terminal UI and CLI design patterns for product owners. Load when designing command-line tools, interactive terminal applications, or writing CLI-specific acceptance criteria. user-invocable: false disable-model-invocation: true
Actionable terminal interface patterns for requirements gathering and design review. Use when target is CLI tool or interactive terminal application.
Follow `program subcommand [flags] [arguments]` (e.g., `git commit -m "message"`).
Required arguments are positional; optional use flags. Accept stdin for piping. Support glob patterns for file arguments. Provide `--dry-run` for destructive/complex operations.
| Framework | Language | Architecture | Best For | |-----------|----------|-------------|----------| | **Bubble Tea** | Go | Elm (Model-Update-View) | Full TUI apps with complex state | | **Rich** | Python | Declarative rendering | Beautiful output, progress, tables | | **Ink** | JavaScript | React component model | Teams with React experience | | **Textual** | Python | CSS-like styling, widgets | Dashboard-style applications | | **Ratatui** | Rust | Immediate-mode rendering | High-performance terminal apps |
Unidirectional data flow makes TUI state predictable and testable.
Arrow keys to navigate, Enter to select. Type-ahead filtering for long lists. Multi-select with space bar, confirm with Enter. Show selected count with clear visual indicator. Support Esc to cancel.
**Red**: errors, failures, deletions | **Yellow/amber**: warnings, attention | **Green**: success, confirmations, additions | **Blue/cyan**: information, links, highlights | **Dim/gray**: secondary info, metadata
Bold for emphasis and headings | Tables for structured data | Indentation for hierarchy | Horizontal rules for section breaks | Unicode box-drawing for visual structure (provide ASCII fallback)
Every CLI error answers three questions:
1. **What happened?** Clear, jargon-free description 2. **Why?** The cause or context 3. **What to do?** Concrete next step
Error: Could not connect to database at localhost:5432
The connection was refused. The database server may not be
running or is not accepting connections on this port.
Try:
1. Check if PostgreSQL is running: pg_isready -h localhost -p 5432
2. Start the server: sudo systemctl start postgresql
3. Verify the port in config: cat ~/.config/myapp/database.yaml
Documentation: https://docs.myapp.com/troubleshooting/databaseError: ECONNREFUSED
tool-name - One-line description USAGE tool-name <command> [flags] COMMANDS create Create a new resource list List existing resources delete Remove a resource FLAGS -h, --help Show this help message -v, --version Show version information -q, --quiet Suppress non-error output EXAMPLES # Create a new project tool-name create my-project --template=web # List all projects tool-name list --format=table LEARN MORE Documentation: https://docs.tool-name.com Report issues: https://github.com/org/tool-name/issues
Lead with examples (users scan for examples first). Show common commands/flags first. One line per description. Link to detailed docs. Support both `tool help <cmd>` and `tool <cmd> --help`.
Detect TTY: terminal gets colors, tables, progress bars. Disable animations/spinners in non-TTY mode. Use structured formatting (headers, indentation, alignment).
Once `--json` schema published, treat as API contract. Breaking changes require major version bump or `--output-version` flag.
AI agents that guide you from idea to working code, with human judgment at every gate. nWave runs inside Claude Code. It breaks feature delivery into seven waves (discover, diverge, discuss, design, devops, distill, deliver).
Repo: nWave-ai/nWave
Review dimensions for validating agent quality - template compliance, safety, testing, and priority validation
Review dimensions for validating agent quality - template compliance, safety, testing, and priority validation
Review dimensions for acceptance test quality - happy path bias, GWT compliance, business language purity, coverage completeness, walking skeleton…
Detailed 5-phase workflow for creating agents - from requirements analysis through validation and iterative refinement
5-layer testing approach for agent validation including adversarial testing, security validation, and prompt injection resistance
Architectural style selection decision matrices, trade-off analysis, structural enforcement rules, and combination patterns. Load when choosing or evaluating…