zai-cli
Execute z.AI CLI providing vision, search, reader, and GitHub exploration via CLI and MCP. Use when user needs image/video analysis, OCR, UI-to-code conversion, error diagnosis, real-time web search, web page to markdown extraction, or GitHub code exploration. Trigger with phrases like "analyze this image", "search the web for", "read this page", "explore this repo", or "use zai". Requires Z_AI_API_KEY.
69
63%
Does it follow best practices?
Impact
Pending
No eval scenarios have been run
Advisory
Suggest reviewing before use
Optimize this skill with Tessl
npx tessl skill review --optimize ./plugins/community/zai-cli/skills/zai-cli/SKILL.mdQuality
Discovery
92%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 description that clearly articulates multiple concrete capabilities, provides explicit trigger phrases, and includes both 'what' and 'when' guidance. Its main weakness is that it bundles many diverse capabilities (vision, search, reading, GitHub) which could overlap with more specialized skills in a large skill library. The inclusion of the tool name 'zai' and API key requirement helps with distinctiveness.
| Dimension | Reasoning | Score |
|---|---|---|
Specificity | Lists multiple specific concrete actions: image/video analysis, OCR, UI-to-code conversion, error diagnosis, real-time web search, web page to markdown extraction, and GitHub code exploration. These are clearly defined capabilities. | 3 / 3 |
Completeness | Clearly answers both 'what' (vision, search, reader, GitHub exploration via CLI/MCP with specific capabilities listed) and 'when' (explicit 'Use when' clause with trigger scenarios and a 'Trigger with phrases like' section). Also notes the API key requirement. | 3 / 3 |
Trigger Term Quality | Includes natural trigger phrases users would actually say: 'analyze this image', 'search the web for', 'read this page', 'explore this repo', 'use zai'. Also includes domain terms like OCR, UI-to-code, vision, and search that users would naturally mention. | 3 / 3 |
Distinctiveness Conflict Risk | While the z.AI CLI branding and 'use zai' trigger are distinctive, several capabilities like 'web search', 'image analysis', and 'read this page' could overlap with other skills that provide similar functionality (e.g., a general web search skill or an image analysis skill). The bundling of many diverse capabilities into one description increases potential overlap. | 2 / 3 |
Total | 11 / 12 Passed |
Implementation
35%Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.
This skill provides a reasonable structural skeleton with setup steps and references to external files, but critically lacks actionable content in the body itself. There are no concrete command examples, no sample inputs/outputs, and the referenced bundle files (errors.md, examples.md) are not provided, leaving Claude with only generic 'use --help' guidance. The skill reads more like a README than an operational instruction set.
Suggestions
Add at least one concrete, copy-paste-ready example for each major capability (vision, search, read, repo) directly in the SKILL.md body, e.g., `npx zai-cli vision analyze --image ./screenshot.png`
Include sample output for at least one command so Claude knows what to expect and can verify success
Provide the referenced bundle files (references/errors.md and references/examples.md) or inline the most critical error handling patterns and examples directly in the skill body
Remove generic filler like 'The CLI is self-documenting - use --help at any level' and replace with specific flag/option guidance for the most common use cases
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The overview paragraph restates what Claude can infer from the command names and description. The prerequisites section is reasonable but 'Network access to Z.AI API endpoints' is obvious padding. The instructions section has some filler ('Try basic commands like vision, search, read, or repo') that doesn't add actionable value. | 2 / 3 |
Actionability | The skill provides no concrete executable examples in the body itself—just generic instructions like 'use --help' and 'try basic commands.' There are no actual command invocations with arguments, no sample inputs/outputs, and the real examples are deferred entirely to a referenced file that isn't provided in the bundle. | 1 / 3 |
Workflow Clarity | The setup steps (1-6) are sequenced and include a validation step ('npx zai-cli doctor'), which is good. However, there's no error recovery guidance inline (just a reference to an external file), and the actual usage workflows for vision/search/read/repo tasks are completely absent—Claude wouldn't know how to chain these commands for real tasks. | 2 / 3 |
Progressive Disclosure | References to errors.md and examples.md are well-signaled and one level deep, which is good structure. However, no bundle files were provided, so the referenced files don't actually exist, making the references dead ends. The main body is also too thin—it offloads nearly all substantive content to files that aren't available. | 2 / 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 |
|---|---|---|
allowed_tools_field | 'allowed-tools' contains unusual tool name(s) | Warning |
frontmatter_unknown_keys | Unknown frontmatter key(s) found; consider removing or moving to metadata | Warning |
Total | 9 / 11 Passed | |
- Commit
3a2d27d
- 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.