zai-cli
Z.AI CLI providing: - Vision: image/video analysis, OCR, UI-to-code, error diagnosis (GLM-4.6V) - Search: real-time web search with domain/recency filtering - Reader: web page to markdown extraction - Repo: GitHub code search and reading via ZRead - Tools: MCP tool discovery and raw calls - Code: TypeScript tool chaining Use for visual content analysis, web search, page reading, or GitHub exploration. Requires Z_AI_API_KEY.
69
84%
Does it follow best practices?
Impact
—
No eval scenarios have been run
Advisory
Suggest reviewing before use
Quality
Discovery
77%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 well-structured with clear capability categories and an explicit 'Use for' clause. Its main weakness is the heavy use of product-specific jargon (GLM-4.6V, ZRead, MCP, Z.AI) that users wouldn't naturally use in requests, and its broad scope covering many domains increases potential overlap with more focused skills.
Suggestions
Add more natural user-facing trigger terms alongside technical ones, e.g., 'analyze screenshot', 'read webpage content', 'search the internet' to improve matching with how users actually phrase requests.
Consider narrowing the 'Use when' clause to emphasize what makes this skill unique compared to other vision, search, or GitHub skills — perhaps emphasizing the Z.AI integration or multi-modal chaining aspect.
| Dimension | Reasoning | Score |
|---|---|---|
Specificity | Lists multiple specific concrete actions organized by category: image/video analysis, OCR, UI-to-code, error diagnosis, web search with filtering, web page to markdown extraction, GitHub code search, MCP tool discovery, and TypeScript tool chaining. | 3 / 3 |
Completeness | Clearly answers both 'what' (detailed capability list across six categories) and 'when' ('Use for visual content analysis, web search, page reading, or GitHub exploration'). The explicit 'Use for...' clause provides clear trigger guidance. | 3 / 3 |
Trigger Term Quality | Includes some natural keywords like 'image', 'video', 'OCR', 'web search', 'GitHub', 'markdown', but many terms are technical/product-specific (GLM-4.6V, ZRead, MCP, Z.AI) that users wouldn't naturally say. Missing common user phrases like 'analyze screenshot', 'read webpage', 'search the web'. | 2 / 3 |
Distinctiveness Conflict Risk | While the Z.AI branding and specific model references (GLM-4.6V) help distinguish it, the broad scope covering vision, search, reading, and code could overlap with dedicated image analysis, web search, or GitHub skills. The combination of many capabilities makes it harder to distinguish from more focused skills. | 2 / 3 |
Total | 10 / 12 Passed |
Implementation
92%Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.
This is a well-crafted CLI skill that is concise, actionable, and clearly structured. The command table provides excellent at-a-glance navigation, and the Quick Start examples are immediately executable. The only weakness is the unverifiable reference to `references/advanced.md` since no bundle files were provided.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The content is lean and efficient. It uses a table for command overview, provides concise examples, and avoids explaining what CLI tools are or how npx works. Every section earns its place. | 3 / 3 |
Actionability | Provides fully executable, copy-paste ready commands for every major feature. The Quick Start section covers all key use cases with real command-line examples including flags and arguments. | 3 / 3 |
Workflow Clarity | For a CLI tool skill, the workflow is straightforward: setup (env var) → run commands → check output. The `doctor` command serves as a validation step. There are no destructive or batch operations requiring feedback loops, so the single-task clarity is sufficient. | 3 / 3 |
Progressive Disclosure | References `references/advanced.md` for advanced topics which is good one-level-deep disclosure, but no bundle files were provided to verify the reference exists. The skill also mentions `--help` as self-documenting which is a reasonable form of progressive disclosure, but the advanced reference path cannot be validated. | 2 / 3 |
Total | 11 / 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
- numman-ali/n-skills
- Commit
a518c67
- 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.