cowart-image-edit

Generate revised AI images from user-supplied Cowart annotation screenshots and place each result beside its original without replacing, moving, hiding, or deleting existing images or annotations.

Quality

66%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Low

Low-risk findings worth noting

Fix and improve this skill with Tessl

tessl review fix ./skills/cowart-image-edit/SKILL.md

Quality

Content

73%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.

The body delivers a clear, well-sequenced workflow with concrete MCP integration details and a proper validation step. The main weakness is verbosity in guardrails and conditional placement rules that could be tightened for token efficiency.

Suggestions

Consolidate the Guardrails section with the inline placement rules in step 6 to remove restated constraints and reduce token weight.

Tighten the image-generation path resolution into a single ordered fallback list rather than prose with caveats.

Move the full JSON MCP call shape and page-asset path details to a reference file if bundle support is added, keeping the body as the overview.

Dimension	Reasoning	Score
Conciseness	Mostly efficient and avoids explaining concepts Claude already knows, but the guardrails, edge-case placement rules, and multiple fallback resolution bullets add padding that could be tightened. Not a 4 because several sections restate constraints already implied by the workflow steps.	3 / 5
Actionability	Provides a concrete MCP call shape with specific fields, a timestamped filename pattern, and explicit placement arithmetic (margin 40, move right by anchor width + 40). Not a 5 because the image-generation path resolution is conditional/probabilistic ('when you can prove…') rather than copy-paste executable.	4 / 5
Workflow Clarity	An explicit 8-step sequence with a dedicated verification step (step 8), fallback/feedback guidance (stale-file resolution, MCP-unavailable fallback), and a visual-confirm checklist. This matches the anchor for clear sequence with explicit validation and error-recovery feedback loops.	5 / 5
Progressive Disclosure	Well-organized into Preconditions, Workflow, and Guardrails with no nested or buried references; as a single self-contained file with no bundle assets, structure is appropriate. Not a 5 because the body inlines material (full JSON call shape, asset-path details) that is borderline detail-level.	4 / 5
	Total	16 / 20 Passed

Description

58%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.

The description states a clear 'what' with concrete, third-person actions and a distinct Cowart-specific niche, but it lacks any explicit 'Use when…' trigger guidance and only partially covers natural user phrasing.

Suggestions

Append a 'Use when…' clause naming the trigger scenario, e.g. 'Use when the user provides a Cowart 批注 screenshot and asks to revise or regenerate the image.'

Add natural user-facing synonyms ('revise', 'redo', 'edit this image', 'regenerate') alongside 'annotation screenshots' to improve trigger term coverage.

Lead with the narrowest discriminator (revising from 批注 screenshots specifically) to push distinctiveness toward a clear niche with minimal overlap.

Dimension	Reasoning	Score
Specificity	Names concrete actions — 'Generate revised AI images' and 'place each result beside its original' — with the protect-existing-content constraint, giving several specific actions. Not a 5 because the action list is narrow (generate + place) rather than comprehensive.	4 / 5
Completeness	The 'what' is clearly stated but there is no 'Use when…' clause or equivalent trigger guidance; per the judging guidelines a missing explicit trigger caps completeness at 3.	3 / 5
Trigger Term Quality	Includes 'annotation screenshots' as a relevant term but misses the natural phrases and synonyms a user would actually say (e.g. 'edit image', 'revise', 'redo'), so keyword coverage is partial.	3 / 5
Distinctiveness Conflict Risk	The Cowart annotation-screenshot niche is distinct with minimal conflict risk; not a 5 because 'generate/revise AI images' could overlap with a generic image-generation skill absent the screenshot qualifier.	4 / 5
	Total	14 / 20 Passed

Validation

100%

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 — 16 / 16 Passed

Validation for skill structure

No warnings or errors.

Repository: zhongerxin/Cowart
Path: skills/cowart-image-edit/SKILL.md
Commit: 43fc888

Reviewed: 5 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.