ccw-cli-tools
CLI tools execution specification (gemini/claude/codex/qwen/opencode) with unified prompt template, mode options, and auto-invoke triggers for code analysis and implementation tasks. Supports configurable CLI endpoints for analysis, write, and review modes.
38
37%
Does it follow best practices?
Impact
—
No eval scenarios have been run
Passed
No known issues
Optimize this skill with Tessl
npx tessl skill review --optimize ./.codex/skills/ccw-cli-tools/SKILL.mdQuality
Discovery
32%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 overly technical and architecture-focused, reading more like an internal specification than a skill selection guide. It names specific CLI tools which helps with distinctiveness, but lacks a clear 'Use when...' clause and natural user-facing trigger terms. The jargon-heavy language ('unified prompt template', 'auto-invoke triggers', 'configurable CLI endpoints') would not help Claude match this skill to typical user requests.
Suggestions
Add an explicit 'Use when...' clause, e.g., 'Use when the user asks to run gemini, claude, codex, qwen, or opencode CLI tools, or wants to analyze, write, or review code using external AI models.'
Replace architectural jargon with user-facing language: instead of 'unified prompt template' and 'configurable CLI endpoints', describe what the user experiences, e.g., 'Runs external AI CLI tools to analyze code, generate implementations, or perform code reviews.'
Add natural trigger terms users would say, such as 'run codex', 'use gemini to review', 'call claude CLI', 'external AI tool', or 'code generation with opencode'.
| Dimension | Reasoning | Score |
|---|---|---|
Specificity | Names the domain (CLI tools execution) and lists some actions (code analysis, implementation tasks, analysis/write/review modes), but the description is heavy on architectural jargon ('unified prompt template', 'configurable CLI endpoints', 'auto-invoke triggers') rather than concrete user-facing actions. | 2 / 3 |
Completeness | The description addresses 'what' (CLI tools execution with modes) but has no explicit 'Use when...' clause or equivalent trigger guidance. The mention of 'auto-invoke triggers' describes an internal mechanism rather than telling Claude when to select this skill. Per rubric guidelines, missing 'Use when...' caps completeness at 2, and the 'when' is so weak it warrants a 1. | 1 / 3 |
Trigger Term Quality | Includes some relevant tool names (gemini, claude, codex, qwen, opencode) which are useful trigger terms, and mentions 'code analysis' and 'implementation tasks'. However, it lacks natural user phrases like 'run CLI', 'execute command', 'code review', or 'analyze code with' that users would actually say. | 2 / 3 |
Distinctiveness Conflict Risk | The specific CLI tool names (gemini, claude, codex, qwen, opencode) provide some distinctiveness, but 'code analysis and implementation tasks' is very broad and could overlap with many coding-related skills. The modes (analysis, write, review) are somewhat distinctive but not sharply delineated. | 2 / 3 |
Total | 7 / 12 Passed |
Implementation
42%Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.
The skill provides highly actionable and concrete CLI execution guidance with excellent examples and a well-structured prompt template system. However, it is severely undermined by extreme verbosity and redundancy — key concepts like configuration loading and mode descriptions are repeated multiple times, inflating the token cost significantly. The monolithic structure with no progressive disclosure means the entire ~400-line document must be loaded into context even for simple tasks.
Suggestions
Eliminate redundancy by consolidating configuration loading instructions into a single section — currently repeated in Initialization, Configuration Reference, and Tool Selection Strategy sections.
Split into multiple files: move the full template catalog to TEMPLATES.md, detailed examples to EXAMPLES.md, and configuration reference to CONFIG.md, keeping SKILL.md as a concise overview with references.
Remove duplicate field descriptions — the configuration fields are listed both as a table and as a bullet list in 'Configuration Fields' section.
Add explicit validation/verification steps after CLI execution — e.g., how to verify output quality, check for error indicators in responses, and criteria for deciding whether to retry with fallback.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | Extremely verbose at ~400+ lines with significant redundancy. The configuration loading instructions are repeated at least 3 times (Initialization, Configuration Reference, Tool Selection Strategy). The configuration fields table is duplicated. Concepts like mode descriptions appear multiple times. Much of this content (JSON parsing, CLI flag usage, what 'read-only' means) assumes Claude doesn't understand basic concepts. | 1 / 3 |
Actionability | Provides fully executable CLI commands with complete prompt templates, concrete examples for multiple task types (security analysis, feature implementation, bug diagnosis, code review), specific flag usage, and copy-paste ready command structures. The 6-field prompt template is well-specified with good and bad examples. | 3 / 3 |
Workflow Clarity | The process flow diagram and decision tree provide clear sequencing, and the fallback chain is well-defined. However, there are no explicit validation checkpoints after execution — no guidance on verifying CLI output quality, checking for errors in the response, or handling malformed output. The 'Handle Results' step is minimal (success → deliver, failure → fallback) without verification criteria. | 2 / 3 |
Progressive Disclosure | This is a monolithic wall of text with no references to external files for detailed content. The rule templates section lists 15+ templates with no links to their definitions. The configuration structure, prompt template details, all examples, and the complete tool selection algorithm are all inline. Content like the full template catalog, detailed examples, and configuration reference would benefit greatly from being split into separate files. No bundle files are provided to support this. | 1 / 3 |
Total | 7 / 12 Passed |
Validation
81%Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.
Validation — 9 / 11 Passed
Validation for skill structure
| Criteria | Description | Result |
|---|---|---|
skill_md_line_count | SKILL.md is long (560 lines); consider splitting into references/ and linking | Warning |
frontmatter_unknown_keys | Unknown frontmatter key(s) found; consider removing or moving to metadata | Warning |
Total | 9 / 11 Passed | |
- Repository
- catlog22/Claude-Code-Workflow
- Commit
5ff5e86
- 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.