create-rule
Use when found gap or repetative issue, that produced by you or implemenataion agent. Esentially use it each time when you say "You absolutly right, I should have done it differently." -> need create rule for this issue so it not appears again.
48
36%
Does it follow best practices?
Impact
Pending
No eval scenarios have been run
Passed
No known issues
Optimize this skill with Tessl
npx tessl skill review --optimize ./plugins/customaize-agent/skills/create-rule/SKILL.mdQuality
Discovery
17%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 description is poorly written with multiple spelling errors, vague language, and informal phrasing that makes it difficult for Claude to reliably select this skill. It fails to specify concrete actions (what kind of rules, where they're stored, what format) and uses no natural trigger keywords that users would employ. The conversational tone ('Esentially use it each time when you say...') is inappropriate for a skill description that needs to function as a precise selection mechanism.
Suggestions
Rewrite with concrete actions: e.g., 'Creates persistent rules in .cursorrules or project configuration files to prevent recurring implementation mistakes. Captures the pattern, root cause, and corrective guidance.'
Add explicit trigger terms users would naturally say: e.g., 'Use when encountering repeated bugs, recurring mistakes, implementation anti-patterns, or when a mistake has been made more than once and needs a preventive rule.'
Fix spelling errors and switch to third-person voice: e.g., 'Generates project-level rules to codify lessons learned from implementation errors' instead of the current informal second-person phrasing.
| Dimension | Reasoning | Score |
|---|---|---|
Specificity | The description uses vague language like 'gap or repetitive issue' and 'create rule for this issue' without specifying concrete actions. It does not list specific capabilities such as what kind of rules are created, where they are stored, or what format they take. | 1 / 3 |
Completeness | The description attempts to answer 'when' with the trigger scenario ('when you say You absolutely right, I should have done it differently') and vaguely addresses 'what' (create a rule to prevent recurrence). However, the 'what' is extremely weak and the 'when' clause is informal and imprecise rather than explicit trigger guidance. | 2 / 3 |
Trigger Term Quality | The description lacks natural keywords a user would say. Terms like 'gap or repetative issue' and 'implemenataion agent' are not phrases users naturally use. It also contains multiple spelling errors ('repetative', 'implemenataion', 'Esentially') which further reduce keyword matching quality. | 1 / 3 |
Distinctiveness Conflict Risk | The description is extremely generic — 'gap or repetitive issue' and 'create rule' could apply to almost any quality improvement, linting, documentation, or process skill. It provides no clear niche or distinct triggers to differentiate it from other skills. | 1 / 3 |
Total | 5 / 12 Passed |
Implementation
55%Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.
The skill provides highly actionable and well-structured workflow guidance for creating rules with contrastive examples, including complete templates, validation checklists, and iterative refinement steps. However, it is severely bloated — explaining concepts Claude already knows (what rules/skills/CLAUDE.md are, why contrastive examples work theoretically), duplicating official documentation verbatim, and cramming everything into a single monolithic file instead of splitting reference material into supporting bundle files.
Suggestions
Cut the 'About Rules' section, the Rules vs Skills vs CLAUDE.md comparison table, and the 'Why Contrastive Examples Work' research explanation — Claude already knows these concepts. This alone would save ~100+ lines.
Remove or move the 'Claude Code Official Rules Guidelines' appendix to a separate reference file — it duplicates official documentation and adds ~60 lines of content Claude can already access.
Extract the complete rule examples and anti-patterns catalog into separate bundle files (e.g., EXAMPLES.md, ANTI-PATTERNS.md) and reference them from the main skill to improve progressive disclosure.
Tighten the 'Writing Effective Rules' section by removing the principle table (these are generic writing advice Claude already follows) and condensing the Incorrect/Correct example guidance into a compact checklist.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The skill is extremely verbose at ~400+ lines. It explains concepts Claude already knows (what rules are, what skills are, what CLAUDE.md is), includes a full comparison table of Rules vs Skills vs CLAUDE.md, explains why contrastive examples work from a research perspective, and even appends the entire official Claude Code rules documentation at the end. Much of this content is redundant or unnecessary for an agent that already understands these concepts. | 1 / 3 |
Actionability | The skill provides fully concrete, executable guidance with complete rule file templates, real code examples (TypeScript), exact file paths, glob patterns, YAML frontmatter examples, and a detailed checklist. The contrastive pattern template and complete rule examples are copy-paste ready. | 3 / 3 |
Workflow Clarity | The 7-step rule creation process is clearly sequenced with explicit validation in Step 6 (file location, frontmatter, contrastive examples, token budget, path scoping, no overlap) and an iterative refinement cycle in Step 7 with decompose, filter, and reweight substeps. The feedback loop (iterate until user confirms rule is good) is explicitly included. | 3 / 3 |
Progressive Disclosure | The entire skill is a monolithic wall of text with no references to external files. Content that could be split out (the complete examples, the anti-patterns catalog, the official Claude Code guidelines appendix, the comparison table) is all inline. For a skill this long (~400+ lines), the lack of any progressive disclosure into supporting files is a significant organizational weakness. | 1 / 3 |
Total | 8 / 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 (612 lines); consider splitting into references/ and linking | Warning |
Total | 10 / 11 Passed | |
- Repository
- NeoLabHQ/context-engineering-kit
- Commit
dedca19
- 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.