agent-writer

Create and validate agent configurations for OpenCode, Gemini CLI, and other AI tools.

2|2|Updated Apr 7, 2013
One-click install
npx skills add https://github.com/warpcode/dotfiles --skill agent-writer
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: agent-writer
Source: https://github.com/warpcode/dotfiles/tree/main/generic/.config/opencode/skills/agent-writer
Command: npx skills add https://github.com/warpcode/dotfiles --skill agent-writer

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

Creates and validates agent configurations for various AI tools including gemini cli, vscode copilot, claude code, cursor, cline, and primarily opencode. Generates new agent configurations from user requirements or reviews existing ones for completeness, clarity, and adherence to best practices. Ensures proper tool access, safety boundaries, and integration with the target AI ecosystem.

Core Features & Use Cases

  • Design and validate agent configurations for Gemini CLI, VSCode Copilot, Claude Code, Cursor, etc.
  • Generate configurations from user requirements or review existing ones for completeness and safety.
  • Ensure tool access, safety boundaries, and integration with the target AI ecosystem.
  • Examples:
    • Context: Creating a new security code review agent for opencode.
    • User: "Design an agent for secure code review with limited tool permissions."
    • Assistant: "I'll use the agent-writer skill to draft a safe, effective configuration."

Quick Start

Trigger: "Create a new agent configuration for opencode."

Frequently Asked Questions about agent-writer

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

FAQPage Schema
How do I create a safe agent configuration for AI tools?

Design agent configurations by defining tool access permissions, safety boundaries, and integration parameters for your target AI platform. Agent-writer validates completeness, clarity, and adherence to best practices, ensuring the configuration operates safely and effectively within your ecosystem.

Can I use agent configurations across multiple AI tools like Claude Code, Cursor, and Cline?

Yes. Agent-writer generates and reviews configurations tailored to multiple AI tools including Claude Code, Cursor, Cline, VSCode Copilot, and Gemini CLI. Each configuration respects the target tool's safety model and integration requirements while maintaining core principles.

What should I check when validating an existing agent configuration?

Validate tool permissions are correctly scoped, safety boundaries are enforced, integration with your AI ecosystem is complete, and the configuration follows established guidelines. Agent-writer reviews these dimensions to catch gaps before deployment.

How do agent configurations ensure safe operation of AI tools?

Configurations define explicit tool access rules, permission scopes, and operational constraints that prevent unsafe behavior. Agent-writer enforces these safety boundaries during design and review, ensuring tools operate within intended limits.

Do I need to design separate configurations for OpenCode versus VSCode Copilot?

Yes. Each AI tool has distinct configuration formats and safety models. Agent-writer generates platform-specific configurations that respect OpenCode, VSCode Copilot, Claude Code, Cursor, Cline, and Gemini CLI requirements independently.

What's the best way to audit an agent configuration for completeness?

Review the configuration against best practices for tool access, safety boundaries, ecosystem integration, and clarity. Agent-writer systematically validates these criteria, identifying missing or misconfigured elements before the agent runs.