delegation-check
Check workflow delegation prompts against agent role definitions for content separation violations. Detects conflicts, duplication, boundary leaks, and missing contracts. Triggers on "check delegation", "delegation conflict", "prompt vs role check".
84
81%
Does it follow best practices?
Impact
Pending
No eval scenarios have been run
Risky
Do not use without reviewing
Quality
Discovery
85%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 well-crafted description that clearly defines a narrow, specific capability around checking delegation prompts against agent role definitions. It covers both what the skill does and when to use it with explicit triggers. The main weakness is that the trigger terms feel somewhat artificial and may not match how users naturally phrase requests in this domain.
Suggestions
Add more natural user language variations as trigger terms, such as 'agent boundaries', 'role overlap', 'responsibility separation', or 'multi-agent prompt review'
| Dimension | Reasoning | Score |
|---|---|---|
Specificity | Lists multiple specific concrete actions: checking delegation prompts against role definitions, detecting conflicts, duplication, boundary leaks, and missing contracts. These are clearly defined capabilities. | 3 / 3 |
Completeness | Clearly answers both what (checks workflow delegation prompts against agent role definitions for content separation violations, detects conflicts/duplication/boundary leaks/missing contracts) and when (explicit triggers on 'check delegation', 'delegation conflict', 'prompt vs role check'). | 3 / 3 |
Trigger Term Quality | Includes some relevant trigger terms like 'check delegation', 'delegation conflict', 'prompt vs role check', but these feel somewhat prescribed rather than natural user language. Missing common variations like 'agent boundaries', 'role overlap', 'responsibility separation'. | 2 / 3 |
Distinctiveness Conflict Risk | Occupies a very specific niche around workflow delegation and agent role boundary checking. The domain is narrow enough that it's unlikely to conflict with other skills, and the trigger terms are highly specific to this use case. | 3 / 3 |
Total | 11 / 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, highly actionable skill for a complex multi-step validation process. Its greatest strength is the concrete, systematic 7-dimension conflict framework with clear severity levels, specific patterns to detect, and explicit exceptions. The main weakness is that the inline detail for all 7 dimensions makes the skill quite long, and some content could be more concise or split into reference files.
Suggestions
Consider moving the detailed 7-dimension check specifications (Section 5) into a separate reference file like `dimensions.md`, keeping only a summary table with dimension names, key questions, and severity levels in the main skill.
Tighten the parsing tables in Steps 3 and 4 — Claude can infer what `<objective>` or `<role>` tags contain without explicit descriptions like 'What to accomplish' or 'Identity, spawner, responsibilities'.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The skill is thorough but verbose in places. Tables explaining block types and agent sections contain information Claude could infer. The 7 dimensions are well-documented but some descriptions repeat patterns (e.g., severity explanations follow the same template). The overall length (~300 lines) is justified by complexity but could be tightened by 20-30%. | 2 / 3 |
Actionability | Highly actionable with concrete grep commands for discovery, specific patterns to match (e.g., 'You are a...', decision tables), structured output formats, and clear severity classifications. Each dimension has explicit check criteria, allowed exceptions, and severity levels. The report template is copy-paste ready. | 3 / 3 |
Workflow Clarity | Excellent 7-step sequential workflow with clear progression from scope determination → discovery → parsing → conflict checks → aggregation → reporting. Includes validation via the 7-dimension checklist, explicit verdicts (CLEAN/REVIEW/CONFLICT), and fix recommendations with before/after examples. The AskUserQuestion fallback for ambiguous scope is a good feedback loop. | 3 / 3 |
Progressive Disclosure | References `separation-rules.md` as required reading (good one-level-deep reference), but the skill itself is monolithic — all 7 conflict dimensions with their full check logic are inline. The dimension details could be split into a reference file with the SKILL.md providing an overview table and linking out for detailed check logic. | 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 |
|---|---|---|
allowed_tools_field | 'allowed-tools' contains unusual tool name(s) | Warning |
Total | 10 / 11 Passed | |
- Repository
- catlog22/Claude-Code-Workflow
- Commit
0f8e801
- 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.