Integrates Claude Code for automated reviews, analysis, and investigation tasks.

Install

mkdir -p .claude/skills/claude-yuiseki && curl -L -o skill.zip "https://agentskills.codes/api/skills/download/12831" && unzip -o skill.zip -d .claude/skills/claude-yuiseki && rm skill.zip

Installs to .claude/skills/claude-yuiseki

Activation

This is the description your AI agent reads to decide when to run this skill — the better it matches your request, the more reliably it fires.

Claude Code CLI(Anthropic)を使用してコードや文言について相談・レビューを行う。 トリガー: "claude", "claudeと相談", "claudeに聞いて", "コードレビュー", "レビューして" 使用場面: (1) 文言・メッセージの検討、(2) コードレビュー、(3) 設計の相談、(4) バグ調査、(5) 解消困難な問題の調査
183 charsno explicit “when” trigger
Intermediate

Key capabilities

  • →Consult Claude Code CLI for code and wording
  • →Perform code reviews
  • →Investigate bugs
  • →Analyze project architecture
  • →Propose refactoring plans
  • →Evaluate UI/UX design

How it works

The skill executes the Claude Code CLI with specific parameters, including a prompt that instructs Claude to provide concrete suggestions without asking further questions, and reports the text output.

Inputs & outputs

You give it
Request for consultation, review, or analysis of code or wording, with an explicit instruction to provide concrete suggestions without further questions
You get back
Text-formatted response from Claude Code CLI with concrete suggestions, fixes, or code examples

When to use claude

  • →Review code
  • →Investigate bugs
  • →Analyze project architecture

About claude

Runs the Claude Code CLI to provide immediate feedback, code reviews, and architectural insights based on project context.

|

When not to use it

  • →When the request requires interactive mode
  • →When permissions should not be skipped
  • →When a specific output format other than text is required without specifying it

Limitations

  • →Requires `cd <project_directory>` to navigate to the target project
  • →Requires `--dangerously-skip-permissions` for automated execution in sandbox environments
  • →Requires `--output-format text` for text output, though it is the default

How it compares

This skill automates interaction with the Claude Code CLI for various development tasks, ensuring direct and actionable feedback without conversational overhead, unlike manual, interactive CLI usage.

Compared to similar skills

claude side by side with the closest alternatives in the catalog.

SkillInstallsUpdatedSafetyDifficulty
claude (this skill)09moReviewIntermediate
error-handling-patterns356moNo flagsIntermediate
codebase-context-extractor41yReviewIntermediate
investigating-code-patterns112moNo flagsBeginner

Try saying

Example prompts that trigger this skill in your AI assistant.

Search skills

Search the agent skills registry