opencode-configuration

Configure OpenCode agents, commands, skills, plugins, tools, and MCP servers.

Updated Mar 12, 2026
One-click install
npx skills add https://github.com/mpsuesser/workspace --skill opencode-configuration-mpsuesser
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: opencode-configuration
Source: https://github.com/mpsuesser/workspace/tree/main/dotconfig/opencode/skills/opencode-configuration
Command: npx skills add https://github.com/mpsuesser/workspace --skill opencode-configuration-mpsuesser

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

Speeds up configuring OpenCode by consolidating guidance on where to put settings and instructions for agents, commands, skills, plugins, custom tools, and MCP servers—so you can extend or troubleshoot projects without guessing.

Core Features & Use Cases

  • OpenCode configuration map: Learn the opencode.json schema, config locations, precedence rules, and merge behavior for reliable setup across user, project, and extension scopes.
  • Extension selection & wiring: Choose the right mechanism (agent vs command vs skill vs tool vs plugin vs MCP server) based on the outcome you want.
  • Permissions and tool behavior: Apply allow/ask/deny permissions, understand wildcard matching, and avoid common “why didn’t it trigger?” issues like skill name mismatches.

Quick Start

Ask an AI to “Show me how to configure OpenCode for this project to load a reusable skill from .opencode/skills, set tool permissions so bash commands are ask-approved, and verify the skill triggers correctly.”

Frequently Asked Questions about opencode-configuration

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

FAQPage Schema
How do I configure OpenCode to load a reusable skill from my project directory?

Configure OpenCode to load skills by placing them in the .opencode/skills directory and ensuring SKILL.md frontmatter is valid with matching skill names. Follow opencode.json schema guidance and config precedence rules to correctly merge settings across user, project, and extension scopes.

What is the difference between agents, commands, skills, and MCP servers in OpenCode?

OpenCode extensions differ by outcome: agents handle autonomous interactions, commands execute specific scripts, skills provide reusable instruction packages, and MCP servers connect external tools. Selecting the right mechanism ensures the correct behavior for your configuration.

Why didn't my OpenCode custom skill trigger when I expected it to?

Skills fail to trigger when SKILL.md frontmatter is invalid or skill names do not match. Verify skill-name matching, check opencode.json schema validation, and review permission rules with wildcard precedence to troubleshoot loading and extension routing issues.

How do I set tool permissions in opencode.json to require approval for bash commands?

Set bash commands to ask-approved by applying allow, ask, and deny permission rules in opencode.json. Understanding wildcard matching and permission precedence ensures tool access rules correctly require approval before executing specified commands.

Can I use JSONC format for opencode.json configuration files?

Yes, OpenCode supports JSONC format for configuration files. Using the opencode.json/JSONC schema guidance allows comments and trailing commas while resolving config locations, ensuring valid schema validation and proper merge behavior across scopes.

What are the config precedence and merge rules for OpenCode settings?

OpenCode config precedence resolves settings across user, project, and extension scopes using specific merge semantics. Following these rules ensures predictable overrides and correct extension routing when combining configurations from multiple opencode.json locations.