crap

Rank functions by CRAP score (complexity × lack of real test coverage) on the current branch, then propose either a refactor or missing tests for the worst offender. Use when the user runs /crap or asks to find risky, complex, poorly-tested code.

Quality

88%

Does it follow best practices?

Impact

—

No eval scenarios have been run

Securityby

Passed

No known issues

Quality

Content

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-crafted, highly actionable skill with an excellent multi-step workflow that includes proper validation checkpoints, error recovery paths, and safety guardrails (characterization tests before refactoring). The main weakness is moderate verbosity in the introductory section explaining CRAP's academic origins and the deviation rationale, which consumes tokens without adding actionable value. The progressive disclosure is reasonable but unverifiable without bundle files.

Suggestions

Trim the 'What CRAP means' section to just the formula and threshold — remove the Savoia history and deviation rationale paragraph, which Claude doesn't need to execute the workflow.

Provide the referenced `detectors.md` and `refactor-playbook.md` bundle files, or note their expected contents so the skill is self-contained.

Dimension	Reasoning	Score
Conciseness	The skill is mostly efficient but includes some unnecessary explanation — the 'What CRAP means' section with the Savoia deviation rationale and history is context Claude doesn't need to perform the task. The formula and threshold are useful, but the academic justification adds tokens without actionable value.	2 / 3
Actionability	Highly actionable with exact CLI commands, specific JSON schemas, exit codes, config file format, and concrete decision logic (cc_axis vs test_axis formulas). The workflow provides copy-paste-ready commands with real flags and file paths throughout.	3 / 3
Workflow Clarity	Excellent multi-step workflow with 9 clearly sequenced steps, explicit validation checkpoints (empty survivors → early exit, cache hit/miss split, exit codes 0/1/2), feedback loops (re-run /crap after characterization tests), and a tests-first guardrail before destructive refactoring. The AskUserQuestion confirmation step before applying changes is a strong safety checkpoint.	3 / 3
Progressive Disclosure	References to `detectors.md` and `refactor-playbook.md` are well-signaled as load-on-demand, but no bundle files were provided to verify they exist. The SKILL.md itself is quite long (~180 lines of dense content) and some sections like the detailed Savoia deviation explanation and the full config YAML could potentially be split out. However, the single-level reference structure is good.	2 / 3
	Total	10 / 12 Passed

Description

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 concisely communicates a specific, well-defined capability. It uses third person voice, lists concrete actions, includes both a command trigger and natural language triggers, and occupies a clear niche that distinguishes it from general code quality or testing skills.

Dimension	Reasoning	Score
Specificity	Lists multiple specific concrete actions: rank functions by CRAP score, calculate complexity × lack of real test coverage, propose refactor or missing tests for the worst offender. These are clear, actionable capabilities.	3 / 3
Completeness	Clearly answers both 'what' (rank functions by CRAP score, propose refactor or missing tests) and 'when' (explicit 'Use when' clause with '/crap' command and natural language triggers like 'find risky, complex, poorly-tested code').	3 / 3
Trigger Term Quality	Includes strong natural trigger terms: '/crap', 'risky', 'complex', 'poorly-tested code', 'CRAP score', 'test coverage'. These cover both the command trigger and natural language variations a user might say.	3 / 3
Distinctiveness Conflict Risk	Highly distinctive with the specific CRAP score methodology, the /crap command trigger, and the unique combination of complexity analysis + test coverage. Unlikely to conflict with general code review or testing skills.	3 / 3
	Total	12 / 12 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 — 11 / 11 Passed

Validation for skill structure

No warnings or errors.

Repository: belchman/claude-skills
Commit: be88d6c

Reviewed: 23 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.