mirror of
https://github.com/IgorWarzocha/Opencode-Workflows.git
synced 2026-09-14 16:22:53 +08:00
OpenCode Agent
2698a1d765
- 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.
33 lines
1.5 KiB
Markdown
33 lines
1.5 KiB
Markdown
---
|
|
description: Create Opencode pack components
|
|
agent: 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".
|