Is this skill user-invocable?

No. It’s designed for orchestrated analyze/debug requests and is invoked via a dedicated subagent (team-shinchan:hiroshi).

What analysis types are performed?

Code Analysis, Bug Analysis, Performance Analysis, and Architecture Analysis are included.

What happens if the input is too long?

Input longer than 2000 characters is truncated to 2000 and a warning is issued.

team-shinchan:analyze

npx machina-cli add skill seokan-jeong/team-shinchan/analyze --openclaw

Files (1)

SKILL.md

1.3 KB

EXECUTE IMMEDIATELY

Step 1: Validate Input

If args is empty or only whitespace:
  Ask user: "What would you like to analyze?"
  STOP and wait for user response

If args length > 2000 characters:
  Truncate to 2000 characters
  Warn user: "Request was truncated to 2000 characters"

Step 2: Execute Task

Do not read further. Execute this Task NOW:

Task(
  subagent_type="team-shinchan:hiroshi",
  model="opus",
  prompt=`/team-shinchan:analyze has been invoked.

## Deep Analysis Request

Perform the following types of analysis:

| Type | Analysis Content |
|------|----------|
| Code Analysis | Structure, dependencies, complexity |
| Bug Analysis | Error cause, stack trace, reproduction conditions |
| Performance Analysis | Bottlenecks, memory, optimization strategies |
| Architecture Analysis | Overall structure, improvements, tradeoffs |

## Result Requirements

- Current state summary
- Issues discovered
- Recommended solutions
- Related file and line references

User request: ${args || '(Please describe what to analyze)'}
`
)

STOP HERE. The above Task handles everything.

Source

git clone https://github.com/seokan-jeong/team-shinchan/blob/main/skills/analyze/SKILL.mdView on GitHub

Overview

team-shinchan:analyze provides a deep, multi-faceted review of code, bugs, performance, and architecture, powered by Hiroshi (Oracle). It supports analyze, debug, and “why isn’t it working” requests, delivering a current-state summary, discovered issues, and actionable fixes with file references.

How This Skill Works

Input is validated: empty input prompts the user; input longer than 2000 chars is truncated with a warning. After validation, the task is executed by a team-shinchan:hiroshi subagent (model opus) using a structured prompt that asks for Code, Bug, Performance, and Architecture analyses, plus recommended solutions and references.

When to Use It

When you need a deep code analysis to understand structure, dependencies, and complexity.
When bugs are hard to reproduce and you need cause, reproduction steps, and stack traces.
When performance bottlenecks or memory usage need identification and optimization plans.
When you’re evaluating the overall architecture for improvements and tradeoffs.
When you want a consolidated set of recommendations with related file references.

Quick Start

Step 1: Provide the code area or description to analyze.
Step 2: Ensure the input is within 2000 characters; truncation will trigger a warning if needed.
Step 3: Run the analysis (team-shinchan:analyze) and review the Hiroshi-generated findings and references.

Best Practices

Define the scope clearly and include the relevant files or modules.
Keep input concise but complete; remember the 2000-character limit to avoid truncation.
Provide reproducible steps for bugs, including environment and version details.
Ask for and expect file and line references in the results.
Review results with an eye toward actionable fixes and tradeoffs.

Example Use Cases

Diagnosing a failing API endpoint by analyzing code path, dependencies, and traces.
Root-cause analysis for a crash with stack trace and reproduction conditions.
Profiling a long-running process to identify memory and CPU bottlenecks.
Architecture review to propose modularization and decoupling strategies.
Refactoring guidance with references to impacted files and tests.

Frequently Asked Questions

Add this skill to your agents