prompt-generator
Generate or convert Claude Code prompt files — command orchestrators, skill files, agent role definitions, or style conversion of existing files. Follows GSD-style content separation with built-in quality gates. Triggers on "create command", "new command", "create skill", "new skill", "create agent", "new agent", "convert command", "convert skill", "convert agent", "prompt generator", "优化".
87
85%
Does it follow best practices?
Impact
Pending
No eval scenarios have been run
Risky
Do not use without reviewing
Quality
Discovery
100%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 strong skill description that clearly defines its scope (generating and converting Claude Code prompt files), lists specific artifact types it handles, and provides an extensive set of explicit trigger terms. It uses proper third-person voice and is concise while being comprehensive. The only minor note is that 'GSD-style content separation' and 'quality gates' are somewhat jargon-heavy, but they serve as useful differentiators for the skill's methodology.
| Dimension | Reasoning | Score |
|---|---|---|
Specificity | Lists multiple specific concrete actions: generate prompt files, convert prompt files, and specifies subtypes (command orchestrators, skill files, agent role definitions, style conversion). Also mentions GSD-style content separation and quality gates. | 3 / 3 |
Completeness | Clearly answers both 'what' (generate or convert Claude Code prompt files with specific types listed) and 'when' (explicit trigger phrases listed with 'Triggers on...' clause). Both dimensions are well-covered. | 3 / 3 |
Trigger Term Quality | Excellent coverage of natural trigger terms explicitly listed: 'create command', 'new command', 'create skill', 'new skill', 'create agent', 'new agent', 'convert command', 'convert skill', 'convert agent', 'prompt generator', and even a Chinese term '优化'. These are phrases users would naturally say. | 3 / 3 |
Distinctiveness Conflict Risk | Highly distinctive — targets a very specific niche of Claude Code prompt file generation/conversion with explicit artifact types (command orchestrators, skill files, agent role definitions). The trigger terms are specific enough to avoid conflicts with general coding or document skills. | 3 / 3 |
Total | 12 / 12 Passed |
Implementation
70%Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.
This is a highly actionable and well-structured orchestration skill with excellent workflow clarity and progressive disclosure through external spec files. Its primary weakness is extreme verbosity — the content could likely be cut by 40-50% without losing any actionable information, as many tables and checks are repetitive across the command/skill/agent variants. The skill would benefit significantly from consolidating shared logic and reducing redundant validation tables.
Suggestions
Consolidate the three nearly-identical quality gate sections (6b, 6b-skill, 6c) into a shared base checklist plus type-specific delta checks, reducing ~60 lines to ~25
Merge the duplicated artifact type detection tables in Steps 1 and 5c.2 into a single reference, or move the detailed classification to the conversion-spec.md
Remove parameter validation tables that Claude can infer (e.g., 'min 10 chars' for description, regex for name) — these add token cost without meaningful guidance
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | This skill is extremely verbose at ~400+ lines. It over-explains routing logic, parameter tables, and validation matrices that could be dramatically condensed. Many tables repeat information (e.g., the artifact type detection table appears twice for convert mode). The quality gate section alone spans dozens of lines with repetitive check tables across command/skill/agent types. | 1 / 3 |
Actionability | The skill provides highly concrete, executable guidance throughout: specific path patterns, exact bash commands, structured Agent() call patterns, precise validation criteria with pass/fail thresholds, and detailed AskUserQuestion examples with complete option structures. Every step has specific actions rather than vague directions. | 3 / 3 |
Workflow Clarity | The 8-step workflow is clearly sequenced with explicit validation checkpoints: Step 6 is a mandatory quality gate with pass/review/fail conditions and retry logic, Step 5c.5 has content loss verification with specific thresholds, Step 7 includes post-write verification, and there are clear feedback loops (fail → re-generate, second fail → ask user). The workflow handles error recovery at multiple points. | 3 / 3 |
Progressive Disclosure | The skill effectively uses progressive disclosure by referencing external spec and template files (command-design-spec.md, agent-design-spec.md, conversion-spec.md, command-md.md, agent-md.md) for detailed design rules while keeping the main file focused on orchestration flow. References are one level deep and clearly signaled with descriptive names. | 3 / 3 |
Total | 10 / 12 Passed |
Validation
90%Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.
Validation — 10 / 11 Passed
Validation for skill structure
| Criteria | Description | Result |
|---|---|---|
allowed_tools_field | 'allowed-tools' contains unusual tool name(s) | Warning |
Total | 10 / 11 Passed | |
- Repository
- catlog22/Claude-Code-Workflow
- Commit
0f8e801
- 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.