security-cleanup-inspector

Removes app bloat, secures data, and ensures failures are visible. Use this skill to safely eliminate dead code or unused resources, secure vulnerabilities like hardcoded API keys, fix silent exceptions (empty catch blocks), convert heavy assets to WebP, and ensure logs do not contain PII.

Quality

83%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

High

Do not use without reviewing

Quality

Content

78%Weight 40%Scale 1-5

Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.

A well-structured, actionable instruction skill with concrete Gradle/logging commands and an explicit verification step, appropriate for its single-purpose cleanup scope. The main weaknesses are mild verbosity from persona decoration and an implicit rather than explicit error-recovery loop in VERIFY.

Suggestions

Trim the persona/philosophy decoration (e.g., the 'Inspector 🕵️' framing and 'Philosophy' bullets) to tighten the token budget.

Make the VERIFY feedback loop explicit: add 'If the build or tests fail, fix the regression and re-run before presenting' so the validate→fix→retry cycle is unambiguous.

Add the exact asset-conversion command or a one-line Gradle snippet for WebP so the WebP step is fully copy-paste ready alongside the other concrete commands.

Dimension	Reasoning	Score
Conciseness	Mostly lean and free of basic-concept over-explanation, but the persona flourish ('The Inspector 🕵️', philosophy bullets) and the dense journaling-rules paragraph add padding that could be trimmed, fitting the 4-anchor rather than the lean 5.	4 / 5
Actionability	Concrete commands and patterns are given ('./gradlew ktfmtFormat', './gradlew lintDebug', 'Timber.e(e)', 'BuildConfig', '.webp') with specific examples, but a few gaps (no exact WebP conversion command or Gradle snippet) keep it just below fully copy-paste ready.	4 / 5
Workflow Clarity	A clear five-step sequence (SCAN → SELECT → SECURE & CLEAN → VERIFY → PRESENT) with real checkpoints (lint confirmation before deleting, build+test in VERIFY), but the error-recovery feedback loop ('if build/tests fail, fix and retry') is implied rather than explicit, matching the 4-anchor.	4 / 5
Progressive Disclosure	A short (~46 line), single-purpose skill with no need for external reference files, organized into clearly labeled sections (Goal, Constraints, Instructions, Examples); per the simple-skill exception this earns a 5, and the only external pointer (.jules/inspector.md) is a one-level runtime journal, not a nested reference.	5 / 5
	Total	17 / 20 Passed

Description

88%Weight 40%Scale 1-5

Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.

A strong, third-person description that pairs a clear capability statement with explicit 'Use this skill to...' trigger guidance and concrete, actionable keywords. It is comprehensive and largely distinct, with only minor room to add synonyms or extensions for trigger coverage.

Dimension	Reasoning	Score
Specificity	Lists multiple concrete actions — 'eliminate dead code or unused resources', 'secure vulnerabilities like hardcoded API keys', 'fix silent exceptions (empty catch blocks)', 'convert heavy assets to WebP', 'ensure logs do not contain PII' — with comprehensive coverage, matching the 5-anchor example.	5 / 5
Completeness	Explicitly answers both 'what' ('Removes app bloat, secures data, and ensures failures are visible') and 'when' ('Use this skill to safely eliminate dead code...') with concrete trigger phrases, matching the 5-anchor.	5 / 5
Trigger Term Quality	Strong natural keywords a developer would actually say ('dead code', 'unused resources', 'hardcoded API keys', 'WebP', 'PII', 'empty catch blocks'), but missing a few common synonyms or file extensions, placing it just below the comprehensive 5-anchor.	4 / 5
Distinctiveness Conflict Risk	The opening 'Removes app bloat, secures data' is somewhat broad, but the concrete triggers (hardcoded API keys, empty catch blocks, WebP) carve a distinct Android-cleanup niche with only minor overlap risk, fitting the 4-anchor better than 5.	4 / 5
	Total	18 / 20 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 — 16 / 16 Passed

Validation for skill structure

No warnings or errors.

Repository: nekomangaorg/Neko
Path: .agents/skills/inspector/SKILL.md
Commit: 9f8d94d

Reviewed: 7 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.