shogun-bloom-config
Interactive wizard: guided questions with multiple-choice options about subscriptions, then outputs a ready-to-paste capability_tiers YAML + fixed agent model assignments. Trigger: "capability_tiers", "bloom config", "routing setup", "set up model routing", "ルーティング設定", "capability_tiers設定", "モデル設定", "サブスク設定", "model routing"
72
88%
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 a specific, niche capability (interactive wizard for generating capability_tiers YAML configuration). It excels in trigger term coverage with both English and Japanese terms, and explicitly separates the 'what' from the 'when' with a dedicated Trigger clause. The description is concise yet comprehensive, making it easy for Claude to select appropriately.
| Dimension | Reasoning | Score |
|---|---|---|
Specificity | Lists multiple concrete actions: 'guided questions with multiple-choice options about subscriptions', 'outputs a ready-to-paste capability_tiers YAML', and 'fixed agent model assignments'. These are specific, actionable capabilities. | 3 / 3 |
Completeness | Clearly answers both 'what' (interactive wizard that guides through subscription questions and outputs capability_tiers YAML + agent model assignments) and 'when' (explicit 'Trigger:' clause with multiple specific trigger terms). | 3 / 3 |
Trigger Term Quality | Excellent coverage of natural trigger terms in both English and Japanese: 'capability_tiers', 'bloom config', 'routing setup', 'set up model routing', 'ルーティング設定', 'model routing', etc. These are terms users would naturally say when needing this skill. | 3 / 3 |
Distinctiveness Conflict Risk | Highly distinctive niche — capability_tiers YAML generation, bloom config, and model routing are very specific domain concepts unlikely to overlap with other skills. The bilingual trigger terms further narrow the scope. | 3 / 3 |
Total | 12 / 12 Passed |
Implementation
77%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 interactive wizard skill with excellent actionability — every path produces copy-paste-ready YAML configs with concrete model names and benchmarks. The workflow sequencing with conditional questions is clearly defined. The main weakness is that all six patterns (with Pattern E having four sub-variants) are inlined in a single file, making it ~400 lines when it could benefit from splitting pattern details into referenced files. Some minor redundancy exists across pattern tables.
Suggestions
Split the six pattern definitions into separate referenced files (e.g., `patterns/pattern-a.md`, `patterns/pattern-e.md`) and keep only the decision matrix, question flow, and Step 5 application instructions in the main SKILL.md.
Consolidate the repeated coverage table format — consider a single template note at the top rather than repeating the same table structure for each pattern.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The skill is quite long (~400 lines) with extensive pattern tables and YAML blocks. While the YAML configs are necessary reference material, there's some redundancy across patterns (repeated table structures, similar coverage tables). The inline comments in YAML are helpful but some explanatory text could be tightened. However, it avoids explaining basic concepts Claude already knows. | 2 / 3 |
Actionability | Highly actionable: every pattern provides copy-paste-ready YAML configs, exact AskUserQuestion call parameters with options, a clear decision matrix mapping answers to patterns, and concrete model names with benchmark scores. The Step 5 application instructions include exact file paths and bash commands. | 3 / 3 |
Workflow Clarity | The workflow is clearly sequenced (Q1→Q2→conditional Q3→pattern mapping→output→application steps) with explicit conditional branching (Q3 only when both subscriptions exist, Q3b only when Q2=Pro). Step 5 includes a validation command (`validate_subscription_coverage`). The decision tree at the end provides a quick mental model. Coverage gap warnings (⚠️ GAP) serve as built-in validation checkpoints. | 3 / 3 |
Progressive Disclosure | All six patterns with their full YAML configs are inline in a single file, making it very long. The Pattern E alone has four sub-variants. This content would benefit from splitting patterns into separate files (e.g., patterns/pattern-e.md) with the SKILL.md serving as an overview with the decision tree and references. However, the section headers and pattern labels do provide reasonable navigation within the monolithic structure. | 2 / 3 |
Total | 10 / 12 Passed |
Validation
90%Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.
Validation — 10 / 11 Passed
Validation for skill structure
| Criteria | Description | Result |
|---|---|---|
skill_md_line_count | SKILL.md is long (518 lines); consider splitting into references/ and linking | Warning |
Total | 10 / 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.