sub-agent-review

Review Claude Code sub-agent configurations for best practices in tools, hooks, and delegation.

8|1|Updated Jan 19, 2026
One-click install
npx skills add https://github.com/igbuend/grimbard --skill sub-agent-review
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: sub-agent-review
Source: https://github.com/igbuend/grimbard/tree/main/skills/sub-agent-review
Command: npx skills add https://github.com/igbuend/grimbard --skill sub-agent-review

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

This Skill streamlines the process of reviewing and optimizing Claude Code sub-agent configurations, ensuring they adhere to best practices for efficiency, security, and functionality.

Core Features & Use Cases

  • Configuration Validation: Checks for correct frontmatter, tool access, model selection, and permission modes.
  • Best Practice Adherence: Evaluates against established guidelines for naming, descriptions, system prompts, and hooks.
  • Use Case: When developing a new AI agent that relies on sub-agents, use this Skill to audit its configuration file before deployment, catching potential issues related to tool permissions or unclear delegation instructions.

Quick Start

Use the sub-agent-review skill to audit the configuration file located at '/path/to/your/agent.md'.

Frequently Asked Questions about sub-agent-review

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I audit Claude Code sub-agent configurations for best practices?

To audit Claude Code sub-agent configurations, you review frontmatter, tool grants, model selection, permission modes, system prompts, and hooks against established design principles. This validates agent configuration for efficiency, security, and functionality before deployment.

What does a sub-agent configuration audit check for?

A sub-agent configuration audit checks frontmatter, tool access permissions, model selection, and hook configurations. It validates naming conventions, descriptions, system prompts, and delegation patterns against established criteria to ensure proper AI agent design.

When do I need to review my AI agent's tool permissions and hooks?

You need to review AI agent tool permissions and hooks before deploying a new sub-agent that relies on delegation. Auditing catches potential issues related to tool access, unclear instructions, and permission modes that could impact security or functionality.

Can I validate Claude Code agent naming and description conventions?

Yes, you can validate Claude Code agent naming and description conventions by evaluating the sub-agent's system prompt and frontmatter against established guidelines. This ensures the overall agent design adheres to best practices for clear delegation instructions.

Does this sub-agent review check model selection and permission modes?

Yes, this sub-agent review checks model selection and permission modes as part of its configuration validation process. It analyzes these settings alongside tool grants and hook configurations to ensure adherence to best practices.