shogun-screenshot
スクリーンショットの取得・加工を行う。ローカルスクショから最新画像を取得、 PlaywrightでWebページをキャプチャ、画像のトリミング・リサイズ、機微情報を黒塗りマスキング。 記事執筆、レポート作成、UI確認、画像加工時に起動。 「スクショ」「スクリーンショット」「画面キャプチャ」「最新のスクショ」「画像加工」「トリミング」「マスク」「写メ」「写メ撮った」「スクショ撮った」で起動。 Do NOT use for: 画像生成（shogun-imagegenを使え）。
67
82%
Does it follow best practices?
Impact
—
No eval scenarios have been run
Advisory
Suggest reviewing before use
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 an excellent skill description that covers all key dimensions thoroughly. It lists specific concrete capabilities, provides extensive natural-language trigger terms (including colloquial Japanese variations), clearly separates 'what' from 'when', and explicitly defines boundaries with other skills via the 'Do NOT use for' clause. The description uses third-person voice appropriately throughout.
| Dimension | Reasoning | Score |
|---|---|---|
Specificity | Lists multiple specific concrete actions: retrieving latest screenshots from local storage, capturing web pages with Playwright, trimming/resizing images, and masking sensitive information with black overlay. | 3 / 3 |
Completeness | Clearly answers both 'what' (screenshot capture, Playwright web capture, trimming/resizing, masking) and 'when' (記事執筆、レポート作成、UI確認、画像加工時に起動) with explicit trigger terms listed. Also includes a 'Do NOT use for' clause to reduce false positives. | 3 / 3 |
Trigger Term Quality | Excellent coverage of natural trigger terms in Japanese including colloquial variations: 「スクショ」「スクリーンショット」「画面キャプチャ」「最新のスクショ」「画像加工」「トリミング」「マスク」「写メ」「写メ撮った」「スクショ撮った」. These are terms users would naturally say. | 3 / 3 |
Distinctiveness Conflict Risk | Clearly carved niche around screenshot capture and image processing with explicit boundary against image generation (pointing to shogun-imagegen). The specific trigger terms and the 'Do NOT use for' clause make it highly distinguishable from other skills. | 3 / 3 |
Total | 12 / 12 Passed |
Implementation
64%Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.
This is a solid, actionable skill with concrete executable commands for all four modes and good use of CLI examples. Its main weaknesses are some unnecessary preamble (North Star, When to Use), missing validation steps in Modes 2 and 3, and a monolithic structure that could benefit from splitting detailed mode instructions into separate files. The masking mode is the strongest section with its preview-then-execute workflow.
Suggestions
Remove or drastically shorten the 'North Star' and 'When to Use' sections — these don't add actionable guidance for Claude.
Add validation/verification steps to Mode 2 (check screenshot file exists and is non-empty after capture) and Mode 3 (verify trimmed output dimensions match expectations).
Consider splitting detailed mode instructions into separate files (e.g., modes/masking.md) and keeping SKILL.md as a concise overview with references.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The 'North Star' section and 'When to Use' section add little actionable value and could be removed. The 'Overview' section is somewhat redundant with the mode descriptions that follow. However, the mode instructions themselves are reasonably efficient with concrete commands. | 2 / 3 |
Actionability | Each mode provides concrete, executable commands with specific script paths, CLI arguments, and options. The masking mode is particularly well-documented with single/multiple region examples and a preview flag. Code is copy-paste ready. | 3 / 3 |
Workflow Clarity | Mode 4 (masking) has a good validation workflow with --preview before final execution. Mode 1 has clear fallback logic. However, Mode 2 (Playwright) and Mode 3 (trimming) lack validation/verification steps — there's no check that the capture succeeded or that the trimmed output is correct. Batch processing is mentioned in guidelines but has no validation guidance. | 2 / 3 |
Progressive Disclosure | Content is well-structured with clear sections per mode, but everything is inline in a single file. The configuration, four full mode descriptions, and guidelines make this fairly long. References to helper scripts (capture_local.sh, trim_image.py, mask_sensitive.py) exist but no bundle files are provided to verify them. Some content (like detailed CLI options) could be split into referenced files. | 2 / 3 |
Total | 9 / 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 | |
- Repository
- yohey-w/multi-agent-shogun
- Commit
bb19915
- 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.