opencode-agents

Define and configure OpenCD agents with permissions, tools, and mode settings.

7|2|Updated Jan 18, 2026
One-click install
npx skills add https://github.com/jal-co/jalco-opencode --skill opencode-agents-jal-co
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: opencode-agents
Source: https://github.com/jal-co/jalco-opencode/tree/main/opencode/.config/opencode/skills/opencode-agents
Command: npx skills add https://github.com/jal-co/jalco-opencode --skill opencode-agents-jal-co

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Defines and configures OpenCode agents with precise permissions, tool access, and mode controls to streamline AI-driven workflows.

Core Features & Use Cases

  • Agent Fundamentals: Location, frontmatter schema, and deployment options for project or global scopes.
  • Permissions & Tools: Flexible permission model (edit, bash, webfetch, skill) to tightly control agent capabilities.
  • Use Cases: Create primary agents for user-facing tasks, subagents for orchestration, and all-mode agents for versatile automation.
  • Examples: Provide design patterns and practical examples for agent creation and policy configuration.

Quick Start

Create a primary agent named security-auditor with read-only permissions and enabled question tool.

Frequently Asked Questions about opencode-agents

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

FAQPage Schema
How do I configure OpenCode agents with specific permissions and tools?

To configure OpenCode agents, you define precise permissions and tool access using a frontmatter schema. This controls capabilities like edit, bash, webfetch, and skill access across project or global scopes.

What is the difference between primary, subagent, and all-mode agents in OpenCode?

Primary agents handle user-facing tasks directly, subagents manage orchestration behind the scenes, and all-mode agents provide versatile automation across every operational state within your OpenCode environment.

Can I restrict an OpenCode agent to read-only permissions?

Yes, you can restrict OpenCode agents to read-only permissions. The flexible permission model allows you to tightly control agent capabilities, limiting access to tools like bash and webfetch for security.

How do I deploy OpenCode agents across project or global scopes?

You deploy OpenCode agents by placing their configuration files in specific directories. This determines whether the agent operates within a local project scope or applies globally across your entire OpenCode setup.

What tools can I enable in an OpenCode agent's frontmatter configuration?

You can enable tools such as bash, webfetch, edit, and skill within an OpenCode agent's frontmatter. The configuration schema allows you to customize these tool access patterns based on your automation needs.

Does OpenCode support creating subagents for task orchestration?

Yes, OpenCode supports creating subagents specifically designed for orchestration. You can configure these subagents with tailored permissions and tools to coordinate complex workflows alongside primary agents.