baoyu-image-gen
[Deprecated: use baoyu-imagine] AI image generation with OpenAI, Azure OpenAI, Google, OpenRouter, DashScope, Z.AI GLM-Image, MiniMax, Jimeng, Seedream and Replicate APIs. Supports text-to-image, reference images, aspect ratios, and batch generation from saved prompt files. Sequential by default; use batch parallel generation when the user already has multiple prompts or wants stable multi-image throughput. Use when user asks to generate, create, or draw images.
72
92%
Does it follow best practices?
Impact
—
No eval scenarios have been run
Passed
No known issues
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 communicates its purpose, supported APIs, capabilities, and when to use it. The deprecation notice is helpful for routing, and the distinction between sequential and batch parallel modes adds useful operational guidance. The only minor concern is that the deprecation notice at the start could cause confusion about whether this skill should be used at all.
| Dimension | Reasoning | Score |
|---|---|---|
Specificity | Lists multiple concrete actions and capabilities: text-to-image, reference images, aspect ratios, batch generation from saved prompt files, sequential vs batch parallel generation. Also enumerates specific API providers. | 3 / 3 |
Completeness | Clearly answers both 'what' (AI image generation with multiple APIs, supporting text-to-image, reference images, aspect ratios, batch generation) and 'when' ('Use when user asks to generate, create, or draw images'). Also includes guidance on sequential vs parallel modes. | 3 / 3 |
Trigger Term Quality | Includes natural trigger terms users would say: 'generate', 'create', 'draw images'. Also mentions specific providers (OpenAI, Google, etc.) and technical terms like 'text-to-image', 'reference images', 'aspect ratios' that users might naturally use. | 3 / 3 |
Distinctiveness Conflict Risk | Highly distinctive with a clear niche: AI image generation via specific API providers. The deprecation notice pointing to 'baoyu-imagine' and the enumeration of specific services (DashScope, Z.AI GLM-Image, MiniMax, Jimeng, Seedream, Replicate) make it very unlikely to conflict with other skills. | 3 / 3 |
Total | 12 / 12 Passed |
Implementation
85%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-structured, highly actionable skill that covers a complex multi-provider image generation system with clear workflows, concrete CLI examples, and excellent progressive disclosure to provider-specific references. Its main weakness is moderate verbosity — several sections (environment variables, dialect explanations, generation mode decision table) could be more concise without losing clarity. Overall it's a strong skill that effectively guides Claude through a complex task.
Suggestions
Consolidate the environment variables table by grouping common patterns (e.g., '<PROVIDER>_API_KEY' and '<PROVIDER>_IMAGE_MODEL' as patterns rather than listing every provider individually) to reduce token count.
Tighten the 'Generation Mode' section — the decision table is useful but the surrounding prose ('Rule of thumb...', 'Use subagents only when...') could be folded into the table's 'Why' column.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The skill is fairly comprehensive but includes some redundancy — environment variables, model resolution, and dialect explanations are quite verbose. The provider selection priority lists and quality preset tables are useful but could be more compact. Some sections like 'Generation Mode' include explanatory prose that could be tightened. | 2 / 3 |
Actionability | The skill provides fully executable CLI commands with concrete flags, clear option tables, and specific examples covering basic, batch, reference image, and provider-specific invocations. The Step 0 blocking setup flow is explicit and actionable with clear file paths and fallback behavior. | 3 / 3 |
Workflow Clarity | The workflow is clearly sequenced: Step 0 (blocking config load) → provider/model resolution → generation with explicit mode selection criteria. Error handling includes auto-retry with attempt counts, and the batch mode includes success/failure reporting. The blocking gate on EXTEND.md is a strong validation checkpoint. | 3 / 3 |
Progressive Disclosure | The skill provides a clear overview with well-organized tables pointing to one-level-deep references for each provider, configuration schema, setup flow, and extended examples. The references table at the bottom provides clean navigation, and provider-specific details are appropriately delegated to separate files. | 3 / 3 |
Total | 11 / 12 Passed |
Validation
72%Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.
Validation — 8 / 11 Passed
Validation for skill structure
| Criteria | Description | Result |
|---|---|---|
metadata_version | 'metadata.version' is missing | Warning |
metadata_field | 'metadata' should map string keys to string values | Warning |
frontmatter_unknown_keys | Unknown frontmatter key(s) found; consider removing or moving to metadata | Warning |
Total | 8 / 11 Passed | |
- Repository
- jimliu/baoyu-skills
- Commit
bd5745f
- 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.