seedream

Generate AI images using Volcengine Seedream model. Supports text-to-image (T2I), image editing (I2I), multi-image fusion, and web-search-based generation. Use this skill when the user wants to create, generate, or edit images.

Quality

76%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Passed

No findings from the security scan

Fix and improve this skill with Tessl

tessl review fix ./SKILLs/seedream/SKILL.md

Quality

Content

65%Weight 40%Scale 1-5

Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.

A highly actionable, copy-paste-ready skill body with excellent concrete examples and a clear sync workflow. Its main weaknesses are verbosity from prompt-best-practices padding and inlined reference content that could be split into separate files, plus absent validation checkpoints for batch generation.

Suggestions

Move the "提示词最佳实践" and "常见使用场景" sections into a separate references/ file (or drop them) since Claude already knows how to write image prompts; keep only a one-line pointer if needed.

Add an explicit validation/verification checkpoint to the sequential/batch workflow (e.g., confirm all N numbered output files exist and are non-empty before declaring success).

Reference the bundled generate-image.ps1 and generate_image.js from the body (or note Windows/Node entry points) so all bundle files are discoverable, and split the parameter/model/error reference into a references/ doc to reduce SKILL.md length.

Dimension	Reasoning	Score
Conciseness	The body is mostly efficient reference material (parameter tables, model list, error catalog), but the "提示词最佳实践" and "常见使用场景" sections re-teach prompt-writing Claude already knows and the API-key setup is duplicated across "快速开始" and "如何配置 API Key", placing it at mostly-efficient-with-some-padding rather than the noticeably-verbose level 2.	3 / 5
Actionability	Provides fully executable, copy-paste-ready bash invocations of the real generate-image.sh script with concrete flags (--prompt, --image, --sequential, --search, --output) covering all common cases (T2I, I2I, multi-image, sequential, search), matching the level-5 anchor.	5 / 5
Workflow Clarity	The sync workflow is clearly sequenced (提交请求 → 等待生成 → 下载图片) with a status table, but batch operations (组图/--sequential) lack explicit validation or verification checkpoints, and the rubric caps batch workflows without validation at 3, so it cannot reach level 4 despite the otherwise clean sequence.	3 / 5
Progressive Disclosure	Section headers are well organized and the entry script generate-image.sh is clearly referenced (and exists in scripts/), but all advanced/reference content (parameters, models, errors, prompt tips) is inlined in one ~440-line SKILL.md with no separate reference .md files, and the bundled .ps1/.js scripts are never referenced from the body, leaving it at some-structure-could-be-better-organized rather than the well-split level 4.	3 / 5
	Total	14 / 20 Passed

Description

88%Weight 40%Scale 1-5

Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.

A strong, well-structured description that concretely lists the supported generation modes and pairs a clear capability statement with an explicit use-when trigger. It is held back only by slightly broad image triggers and a few missing natural synonyms.

Dimension	Reasoning	Score
Specificity	Lists multiple concrete generation modes — "text-to-image (T2I)", "image editing (I2I)", "multi-image fusion", and "web-search-based generation" — giving comprehensive coverage of the skill's capabilities, matching the level-5 anchor rather than 4 which requires minor gaps.	5 / 5
Completeness	Explicitly answers what ("Generate AI images... Supports text-to-image, image editing, multi-image fusion, web-search-based generation") and when ("Use this skill when the user wants to create, generate, or edit images") with concrete trigger phrases, matching the level-5 anchor exactly.	5 / 5
Trigger Term Quality	Includes natural user phrases "create, generate, or edit images" plus "text-to-image" and "image editing", but omits common synonyms like "make", "draw", "pictures", or "photos", so it sits at good-but-not-comprehensive coverage rather than 5.	4 / 5
Distinctiveness Conflict Risk	Naming the specific "Volcengine Seedream model" carves a clear niche, but the trigger "create, generate, or edit images" is broad and could overlap with other image-generation skills, placing it at mostly-distinct with minor overlap risk rather than minimal conflict.	4 / 5
	Total	18 / 20 Passed

Validation

93%

Warnings & errors only

Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.

Validation — 15 / 16 Passed

Validation for skill structure

Criteria	Description	Result
frontmatter_unknown_keys	Unknown frontmatter key(s) found; consider removing or moving to metadata	Warning

	Total	15 / 16 Passed

Repository: netease-youdao/LobsterAI
Path: SKILLs/seedream/SKILL.md
Commit: 100d3f7

Reviewed: 2 days ago

Table of Contents

Discovery Implementation Validation

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.