CtrlK
BlogDocsLog inGet started
Tessl Logo

image-generation

Use this skill when the user requests to generate, create, imagine, or visualize images including characters, scenes, products, or any visual content. Supports structured prompts and reference images for guided generation.

68

Quality

82%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

SecuritybySnyk

Critical

Do not install without reviewing

SKILL.md
Quality
Evals
Security

Quality

Content

65%

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

The body is highly actionable with concrete commands and complete examples, and it is reasonably concise, but it has two concrete defects: a dead reference to templates/doraemon.md (the file is absent) and an internally inconsistent worked example whose Execute block names a different prompt file (cyberpunk-hacker.json) than the one just created (asian-woman.json). Workflow clarity is capped at 2 by the absence of any validation/verification checkpoint.

Suggestions

Create the missing templates/doraemon.md file or remove the dead reference to templates/doraemon.md so progressive_disclosure references resolve to real files.

Fix the Character Generation Example so the Execute block points at the prompt file actually created (asian-woman.json and asian-woman-01.jpg), so the worked example is internally consistent and copy-paste ready.

Add a lightweight verification/feedback step after generation (e.g., confirm the output file exists and is non-empty, then offer iteration) to raise workflow_clarity from 2 to 3.

Consider moving the two long worked JSON examples into separate reference files (e.g., examples/character.json, examples/scene.json) to tighten the SKILL.md body and improve conciseness.

DimensionReasoningScore

Conciseness

The body is mostly efficient and does not explain concepts Claude already knows, but it carries two large inlined JSON examples (~17-line Tokyo character example and ~25-line Star Wars example) plus three bullet-list scenario sections that partly restate fields already shown in the examples, so it could be tightened. It is not a 3 because the worked examples are heavier than needed; it is not a 1 because nothing is padded with elementary explanations.

2 / 3

Actionability

It provides a fully executable bash command, a complete parameter list (--prompt-file, --reference-images, --output-file, --aspect-ratio), real absolute paths, and two complete copy-paste JSON prompt examples, matching the anchor for fully executable commands and copy-paste-ready examples.

3 / 3

Workflow Clarity

Steps 1->2->3 are clearly sequenced with concrete commands, but there is no validation/verification step that the script succeeded, and the worked example is internally inconsistent: it says to create "asian-woman.json" yet the Execute block calls "--prompt-file .../cyberpunk-hacker.json". It is not a 3 because no explicit validation/feedback checkpoint exists; it is not a 1 because the sequence and commands are present.

2 / 3

Progressive Disclosure

There is a well-signaled template reference ("Read the following template file only when matching the user request" -> templates/doraemon.md) but that file does not exist in the bundle, making it a dead reference, and the large worked JSON examples are inlined where separate reference files could carry them. It is not a 3 because the reference is broken and inline heavy content is not split out; it is not a 1 because there is real section structure and only one (one-level) reference rather than deeply nested ones.

2 / 3

Total

9

/

12

Passed

Description

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.

The description is strong: it states concrete actions, uses natural trigger verbs, includes an explicit "Use when" clause, and occupies a distinct niche. It is not the level below because every dimension's top anchor (multiple concrete actions, good trigger coverage, explicit what-and-when, clear niche) is satisfied.

DimensionReasoningScore

Specificity

It names multiple concrete actions ("generate, create, imagine, or visualize") and enumerates concrete content types ("characters, scenes, products, or any visual content") plus "structured prompts and reference images," matching the anchor for multiple specific concrete actions.

3 / 3

Completeness

It explicitly answers both what (image generation with structured prompts and reference images) and when via an explicit "Use this skill when the user requests..." trigger clause, matching the anchor for clearly answering both what AND when with explicit triggers.

3 / 3

Trigger Term Quality

Natural verbs users say ("generate, create, imagine, or visualize images") plus content keywords ("characters, scenes, products") give good coverage of terms a user would actually say, matching the top anchor.

3 / 3

Distinctiveness Conflict Risk

Image generation is a clear niche with distinct visual-content triggers and is unlikely to conflict with file, document, or code skills, matching the top anchor.

3 / 3

Total

12

/

12

Passed

Validation

93%

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

Validation15 / 16 Passed

Validation for skill structure

CriteriaDescriptionResult

relative_links

Relative link issues: 1 missing

Warning

Total

15

/

16

Passed

Repository
bytedance/deer-flow
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.