skills/igorwarzocha__opencode-workflows
1
0
Fork 0
mirror of https://github.com/IgorWarzocha/Opencode-Workflows.git synced 2026-09-14 16:22:53 +08:00
Code Issues Packages Projects Releases Wiki Activity
Files
master
igorwarzocha__opencode-work…/commands/.opencode/command/create-pack.md
T
OpenCode Agent 2698a1d765 feat: comprehensive repository overhaul and standardization 
- Implement internal maintenance suite: repo-maintainer agent and automated audit/sync scripts.
- Add "Opencode Configurator" suite: agent-architect, command-creator, and skill-creator.
- Refactor core commands (/improve, /refactor, /rmslop) to RFC 2119 + XML standards.
- Standardize all agent metadata and structural tagging across the repository.
- Introduce utility commands: /npm for package management and /create-pack for bundling.
- Overhaul README.md and documentation to reflect new workflow patterns.
2026-01-12 18:16:06 +00:00

1.5 KiB
Raw Permalink Blame History

description, agent
description agent
Create Opencode pack components opencode-configurator

You are tasked with creating a complete "Opencode Pack" based on the user's request: "$ARGUMENTS"

A Pack consists of three integrated components:

  1. Agent: The persona and orchestration logic.
  2. Skill: The specialized knowledge, scripts, and instructions.
  3. Command: The user interface/shortcut to trigger the workflow.

Phase 1: Research & Discovery

CRITICAL: You MUST NOT guess.

  1. Analyze the Request: If the user mentions a specific library, tool, or repo, you MUST understand it deeply.
  2. External Research: Use websearch to find the latest documentation, patterns, and best practices for the subject.
  3. Internal Exploration: If this pack is for the current repository, use the task tool (type: explore) to analyze the codebase structure and patterns.

Phase 2: Skill Loading

You MUST load the following skills to ensure architectural compliance:

  1. skill("agent-architect") - To design the agent.
  2. skill("skill-creator") - To structure the skill.
  3. skill("command-creator") - To build the slash command.

Phase 3: Construction

Follow the workflow defined in each skill to build the components.

  • Skill First: Define the capabilities and scripts.
  • Agent Second: Give the agent permission to use the skill (strict whitelist).
  • Command Third: Create the user interface that routes to the agent.

Execution

Start by researching the subject "$ARGUMENTS".

Reference in New Issue View Git Blame Copy Permalink